prax-20201231

UNITED STATES

SECURITIES AND EXCHANGE COMMISSION

WASHINGTON, DC 20549

FORM 10-K

(Mark One)

For the fiscal year ended December 31, 2020

For the transition period from________ to_________

Commission File Number: 001-39620

PRAXIS PRECISION MEDICINES, INC.

(Exact Name of Registrant as Specified in its Charter)

One Broadway, 16th Floor

(Address of principal executive offices) (Zip Code)

Registrant’s telephone number, including area code: 617-300-8460

Securities registered pursuant to Section 12(b) of the Act:

Title of each class Trading Symbol(s) Name of each exchange on which registered

Common Stock, par value $0.0001 per share PRAX The Nasdaq Global Select Market

Indicate by check mark if the registrant is a well-known seasoned issuer, as defined in Rule 405 of the Securities Act. Yes ☐ No ☒

Indicate by check mark if the registrant is not required to file reports pursuant to Section 13 or Section 15(d) of the Act. Yes ☐ No ☒

Indicate by check mark whether the registrant (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding 12 months (or for such shorter period that the registrant was required to file such reports), and (2) has been subject to such filing requirements for the past 90 days. Yes ☒ No ☐

Indicate by check mark whether the registrant has submitted electronically every Interactive Data File required to be submitted pursuant to Rule 405 of Regulation S-T (§232.405 of this chapter) during the preceding 12 months (or for such shorter period that the registrant was required to submit such files). Yes ☒ No ☐

Indicate by check mark whether the registrant is a large accelerated filer, an accelerated filer, a non-accelerated filer, smaller reporting company, or an emerging growth company. See the definitions of “large accelerated filer,” “accelerated filer,” “smaller reporting company,” and “emerging growth company” in Rule 12b-2 of the Exchange Act.

Large accelerated filer ☐ Accelerated filer ☐

Non-accelerated filer ☒ Smaller reporting company ☒

Emerging growth company ☒

If an emerging growth company, indicate by check mark if the registrant has elected not to use the extended transition period for complying with any new or revised financial accounting standards provided pursuant to Section 13(a) of the Exchange Act. ☐

Indicate by check mark whether the registrant has filed a report on and attestation to its management’s assessment of the effectiveness of its internal control over financial reporting under Section 404(b) of the Sarbanes-Oxley Act (15 U.S.C. 7262(b)) by the registered public accounting firm that prepared or issued its audit report. ☐

Indicate by check mark whether the registrant is a shell company (as defined in Rule 12b-2 of the Exchange Act). Yes ☐ No ☒

The aggregate market value of common stock held by non-affiliates of the registrant, based on the closing price of a share of common stock on October 16, 2020 as reported by the Nasdaq Global Select Market on such date was approximately $731.4 million. The registrant has elected to use October 16, 2020, which was the initial trading date on the Nasdaq Global Select Market, as the calculation date because on June 30, 2020 (the last business day of the registrant’s most recently completed second fiscal quarter) the registrant was a privately held company. This calculation does not reflect a determination that certain persons are affiliates of the registrant for any other purpose.

As of March 1, 2021, the registrant had 38,579,115 shares of common stock, $0.0001 par value per share, outstanding.

Table of Contents

TABLE OF CONTENTS

Part I

Item 1. Business 7

Item 1A. Risk Factors 67

Item 1B. Unresolved Staff Comments 121

Item 2. Properties 122

Item 3. Legal Proceedings 122

Item 4. Mine Safety Disclosures 122

Part II

Item 6. Selected Financial Data 124

Item 7A. Quantitative and Qualitative Disclosures About Market Risk 138

Item 8. Financial Statements and Supplementary Data 139

Item 9A. Controls and Procedures 171

Item 9B. Other Information 171

Part III

Item 10. Directors, Executive Officers and Corporate Governance 172

Item 11. Executive Compensation 177

Item 14. Principal Accounting Fees and Services 196

Part IV

Item 15. Exhibits, Financial Statement Schedules 198

Signatures

Table of Contents

SUMMARY OF THE MATERIAL AND OTHER RISKS ASSOCIATED WITH OUR BUSINESS

•We are a clinical-stage biopharmaceutical company and we have incurred significant losses since our inception. We anticipate that we will continue to incur significant losses for the foreseeable future.

•We will need substantial additional funding, and if we are unable to raise capital when needed, we could be forced to delay, reduce or eliminate our product discovery and development programs or commercialization efforts.

•Our business substantially depends upon the successful development of PRAX-114, PRAX-944 and PRAX-562. If we are unable to obtain regulatory approval for, and successfully commercialize, PRAX-114, PRAX-944 or PRAX-562, our business may be materially harmed.

•Clinical development involves a lengthy, complex and expensive process, with an uncertain outcome. The outcome of preclinical testing and early clinical trials may not be predictive of the success of later clinical trials, and the results of our clinical trials, which to date have primarily been conducted in Australia and New Zealand, may not satisfy the requirements of the FDA or comparable foreign regulatory authorities.

•Our product candidates may cause undesirable side effects or have other properties that could delay or prevent their regulatory approval, limit their commercial potential or result in significant negative consequences following regulatory approval, if obtained.

•The markets for PRAX-114 for major depressive disorder and perimenopausal disorder, PRAX-944 for essential tremor, PRAX-562 for multiple rare neurological conditions and any other product candidates we may develop may be smaller than we expect.

•We face significant competition in an environment of rapid technological and scientific change, and there is a possibility that our competitors may achieve regulatory approval before us or develop therapies that are safer, more advanced or more effective than ours, which may negatively impact our ability to successfully market or commercialize any product candidates we may develop and ultimately harm our financial condition.

•Our success depends in part on our ability to protect our intellectual property. It is difficult and costly to protect our proprietary rights and technology, and we may not be able to ensure their protection.

•We have entered into, and may enter into, license or other collaboration agreements that impose certain obligations on us. If we fail to comply with our obligations under such agreements with third parties, we could lose license rights that may be important to our business.

•Third-party claims of intellectual property infringement may prevent or delay our product discovery and development efforts.

•We expect to depend on collaborations with third parties for the research, development and commercialization of certain of the product candidates we may develop. If any such collaborations are not successful, we may not be able to realize the market potential of those product candidates.

•Business interruptions resulting from COVID-19 or a similar pandemic, epidemic or outbreak of an infectious disease in the United States or worldwide may adversely affect our business.

•The price of our stock may be volatile, and you could lose all or part of your investment.

Table of Contents

SPECIAL NOTE REGARDING FORWARD-LOOKING STATEMENTS

This Annual Report on Form 10-K contains express or implied forward-looking statements that are based on our management’s belief and assumptions and on information currently available to our management. Although we believe that the expectations reflected in these forward-looking statements are reasonable, these statements relate to future events or our future operational or financial performance, and involve known and unknown risks, uncertainties and other factors that may cause our actual results, performance or achievements to be materially different from any future results, performance or achievements expressed or implied by these forward-looking statements. Forward-looking statements contained in this Annual Report on Form 10-K include, but are not limited to, statements about:

•the success, cost and timing of our product development activities and clinical trials;

•our expectations regarding our ability to obtain and maintain intellectual property protection for our product candidates;

•the ability to license additional intellectual property relating to our product candidates from third parties and to comply with our existing license agreements and collaboration agreements;

•the ability and willingness of our third-party research institution collaborators to continue research and development activities relating to our product candidates;

•our ability to commercialize our products in light of the intellectual property rights of others;

•our ability to obtain funding for our operations, including funding necessary to complete further development and commercialization of our product candidates;

•the commercialization of our product candidates, if approved;

•our plans to research, develop and commercialize our product candidates;

•future agreements with third parties in connection with the commercialization of our product candidates and any other approved product;

•the size and growth potential of the markets for our product candidates, and our ability to serve those markets;

•the rate and degree of market acceptance of our product candidates;

•the pricing and reimbursement of our product candidates, if approved;

•regulatory developments in the United States and foreign countries;

•our ability to contract with third-party suppliers and manufacturers and their ability to perform adequately;

•the success of competing therapies that are or may become available;

•our ability to attract and retain key scientific or management personnel;

•the accuracy of our estimates regarding expenses, future revenue, capital requirements and needs for additional financing;

•our expectations regarding the period during which we qualify as an emerging growth company under the JOBS Act, enacted in April 2012, or a smaller reporting company as defined in the Securities Exchange Act of 1934, as amended;

•the development of major public health concerns, including the novel coronavirus outbreak or other pandemics arising globally, and the future impact of it and COVID-19 on our clinical trials, business operations and funding requirements; and

Table of Contents

•other risks and uncertainties, including those listed under the caption “Risk Factors.”

In some cases, you can identify forward-looking statements by terminology such as “may,” “should,” “expects,” “intends,” “plans,” “anticipates,” “believes,” “estimates,” “predicts,” “potential,” “continue” or the negative of these terms or other comparable terminology. These statements are only predictions. You should not place undue reliance on forward-looking statements because they involve known and unknown risks, uncertainties, and other factors, which are, in some cases, beyond our control and which could materially affect results. Factors that may cause actual results to differ materially from current expectations include, among other things, those listed under the section titled “Risk Factors” and elsewhere in this Annual Report on Form 10-K. If one or more of these risks or uncertainties occur, or if our underlying assumptions prove to be incorrect, actual events or results may vary significantly from those implied or projected by the forward-looking statements. No forward-looking statement is a guarantee of future performance. You should read this Annual Report on Form 10-K and the documents that we reference in this Annual Report on Form 10-K and have filed with the Securities and Exchange Commission as exhibits hereto completely and with the understanding that our actual future results may be materially different from any future results expressed or implied by these forward-looking statements.

The forward-looking statements in this Annual Report on Form 10-K represent our views as of the date of this Annual Report on Form 10-K. We anticipate that subsequent events and developments will cause our views to change. However, while we may elect to update these forward-looking statements at some point in the future, we have no current intention of doing so except to the extent required by applicable law. You should therefore not rely on these forward-looking statements as representing our views as of any date subsequent to the date of this Annual Report on Form 10-K.

This Annual Report on Form 10-K also contains estimates, projections and other information concerning our industry, our business and the markets for our product candidates. Information that is based on estimates, forecasts, projections, market research or similar methodologies is inherently subject to uncertainties and actual events or circumstances may differ materially from events and circumstances that are assumed in this information. Unless otherwise expressly stated, we obtained this industry, business, market, and other data from our own internal estimates and research as well as from reports, research surveys, studies, and similar data prepared by market research firms and other third parties, industry, medical and general publications, government data and similar sources. While we are not aware of any misstatements regarding any third-party information presented in this Annual Report on Form 10-K, their estimates, in particular, as they relate to projections, involve numerous assumptions, are subject to risks and uncertainties and are subject to change based on various factors, including those discussed under the section titled “Risk Factors” and elsewhere in this Annual Report on Form 10-K.

Table of Contents

PART I

Item 1. Business

BUSINESS

Company Overview

We are a clinical-stage biopharmaceutical company translating genetic insights into the development of therapies for central nervous system, or CNS, disorders characterized by neuronal imbalance. Normal brain function requires a delicate balance of excitation and inhibition in neuronal circuits, which, when dysregulated, can lead to abnormal function and disease. We are applying insights from genetic epilepsies to broader neurological and psychiatric disorders, using our understanding of shared biological targets and circuits in the brain. We apply a deliberate and pragmatic precision approach, leveraging a suite of translational tools including novel transgenic and predictive translational animal models and electrophysiology markers, to enable an efficient path to proof-of-concept in patients. Through this approach, we have established a broad portfolio, including multiple disclosed programs across CNS disorders, including depression, epilepsy, movement disorders and pain syndromes, with three clinical-stage product candidates. We expect multiple topline clinical trial readouts from all three programs in the next year and anticipate the launch of a new clinical development program in 2021. We intend to develop differentiated therapies that can deliver long-term benefits to human health by meaningfully impacting patients and society.

Our most advanced clinical program, PRAX-114, is an extrasynaptic GABAA receptor preferring positive allosteric modulator, or PAM, for the treatment of patients suffering from major depressive disorder, or MDD, and perimenopausal depression, or PMD. Together, these conditions affect more than 22 million people in the United States, many of whom are not responsive to or are underserved by current treatments. PRAX-114 is under development as a potentially differentiated treatment for a broad MDD population, as both a monotherapy and adjunctive therapy for both acute and maintenance treatment. We believe that PRAX-114 has several advantages relative to currently available therapies and product candidates in the GABAA PAM therapeutic class, including the potential for rapid and durable antidepressant effect across MDD symptoms, a wider therapeutic window, simple nightly dosing with or without a meal via tablet formulation and indication expansion opportunities. We have a multi-cohort, three-part Phase 2a clinical trial ongoing in Australia. Parts A and C of the trial are treating patients with MDD while Part B has focused on patients with PMD. For all parts of the trial, PRAX-114 was generally well-tolerated. In Parts A and C, we observed marked improvements in depression scores in MDD patients within two weeks of treatment that were maintained throughout the treatment period. We expect complete topline data from Part B of the trial in the second half of 2021. In October 2020, we submitted an Investigational New Drug application, or IND, to support the initiation of a Phase 2/3 clinical trial in the United States. At the end of the 30-day review period, the U.S. Food and Drug Administration, or the FDA, notified us that the IND was placed on full clinical hold pending the resolution of certain non-clinical pharmacology and toxicology matters. We subsequently interacted with the FDA to gain agreement on a path to initiate the clinical study, which included a proposal to submit available non-clinical data while other good laboratory practice, or GLP, reproductive toxicology studies were being completed. Based on this submission, the FDA removed the clinical hold in March 2021. We are operationally ready and intend to initiate the Phase 2/3 monotherapy MDD trial by the end of March 2021. If positive, the Phase 2/3 trial is intended to serve as one of two registrational trials required by the FDA to support clinical efficacy for monotherapy treatment of MDD, and we expect topline data in the first half of 2022. In addition to the Phase 2/3 monotherapy trial, we intend to initiate a Phase 2 dose range finding, or DRF, trial for adjunctive treatment of MDD in the third quarter of 2021 to provide controlled data to support advancing a Phase 3 adjunctive MDD trial and will further inform dose selection for the future Phase 3 monotherapy trial.

Our second clinical program, PRAX-944, is a potentially differentiated selective small molecule inhibitor of T-type calcium channels for the treatment of Essential Tremor, or ET. ET is a progressive and debilitating movement disorder with action tremor that significantly disrupts activities of daily living, with an estimated prevalence of up to seven million patients in the United States with only one approved pharmacotherapy that is poorly tolerated, resulting in high discontinuation rates. The condition can be debilitating enough that in severe cases, patients opt for invasive brain surgeries when pharmacotherapy fails. Successful development of T-type calcium channel modulators in ET likely requires a pharmacokinetic, or PK, profile with sustained exposure throughout the day and a blunted maximum drug concentration, or Cmax. We believe the therapeutic profile of PRAX-944 coupled with a modified release formulation allows the potential for PRAX-944 to be a differentiated therapy. We have evaluated the safety and tolerability of PRAX-944 in over 150 healthy volunteers in five separate clinical trials. We have studied the safety of PRAX-944 modified release formulation with titration up to 120mg/day and no maximum tolerated dose, or MTD, has been identified. We are currently conducting a Phase 2a proof-of-concept, open-label

Table of Contents

trial, in ET patients. Preliminary site data from six participants in the low dose cohort showed tremor reduction, which compares favorably to the standard of care agents and historical placebo response. Based on the observed safety profile in the healthy volunteer titration study and the safety and preliminary efficacy data in ET participants administered up to 40mg daily, we have added a second cohort to the ongoing ET Phase 2a trial where patients will be titrated to a dose of up to120mg/day of PRAX-944. We have also included a randomized, double-blind, placebo-controlled withdrawal phase to this later cohort in the trial, where participants will either be maintained on their final open-label dose or switched to placebo. We plan to announce topline open-label safety, tolerability and efficacy data, for the high dose cohort, in mid-year 2021. In addition, we plan to start a Phase 1 trial to explore short titration schemes by mid-2021 and to initiate a Phase 2b randomized controlled trial in ET in late 2021.

Our most advanced rare disease product candidate and third clinical program, PRAX-562, is the first selective persistent sodium current blocker in development for the treatment of a broad range of rare, devastating CNS disorders, such as severe pediatric epilepsies and rare adult cephalgias. To date, PRAX-562 has demonstrated pharmacological activity in preclinical in-vivo models at generally well-tolerated doses. We initiated a Phase 1 trial of PRAX-562 in Australia to evaluate the safety, tolerability, PK and effects on an exploratory electroencephalography, or EEG, biomarker in up to 129 adult healthy volunteers. The single ascending dose, or SAD, portion up to the maximum planned dose has been completed with no dose limiting toxicities and the study has advanced to the multiple ascending dose, or MAD, phase. We anticipate initiating the first proof-of-concept trial in patients with rare adult cephalgias, including Short-lasting Unilateral Neuralgiform headache attacks with Conjunctival injection and Tearing, or SUNCT, Short-lasting Unilateral Neuralgiform headache with Autonomic symptoms, or SUNA, and Trigeminal Neuralgia, or TN, in the second half of 2021. The scope of the initial study has been expanded to include TN in addition to SUNCT and SUNA. In January 2021, the FDA granted rare pediatric disease designation for PRAX-562 for the treatment of SCN2A and SCN8A developmental epileptic encephalopathies, or SCN2A-DEE and SCN8A-DEE, respectively.

In addition to our clinical programs, our most advanced preclinical stage program is PRAX-222, an antisense oligonucleotide, or ASO, designed to decrease the expression levels of the protein encoded by the gene SCN2A in patients with gain-of-function, or GOF, mutations in epilepsy. The FDA has granted both rare pediatric disease and orphan drug designations for PRAX-222 for the treatment of SCN2A-DEE. We have one disclosed discovery program in development for KCNT1 related epilepsy and in March 2021 we have entered into an innovative research collaboration with The Florey Institute of Neuroscience and Mental Health to develop three additional novel ASOs for the treatment of patients with severe genetic epilepsies, including a novel approach targeting SCN2A loss-of-function, or LOF mutations.

Below is a summary of our portfolio of disclosed programs, addressing either broad psychiatric and neurologic conditions or rare diseases. We own global commercialization rights for all of our product candidates.

_______________

* PRAX-222 is a collaboration with Ionis Pharmaceuticals, or Ionis, and RogCon Inc. Ionis is eligible to receive royalties as a percentage of net product sales worldwide in the low-20s.

Table of Contents

Our company was founded by scientific innovators Kiran Reddy, M.D., David Goldstein, Ph.D. and Steven Petrou, Ph.D., who have pioneered work to identify and characterize de novo mutations in several dozen genes believed to cause a number of forms of severe pediatric epilepsies. These genes regulate key neuronal circuits in the brain which, when dysregulated, can result in severe seizure phenotypes as well as comorbid developmental delays, cognitive deficits, sensory-motor issues and often early death. Further, based on our understanding of a body of preclinical and clinical evidence, we now believe that these genes also play critical roles in the predisposition to other more prevalent neurologic and psychiatric disorders, such as mood disorders, movement disorders, pain syndromes, autism, migraine and schizophrenia, making them attractive targets for therapeutic intervention for a wide range of CNS disorders.

We have attracted a talented team of scientists and researchers in genetics and biology, chemistry and translational medicine as well as business leaders with established track records of successfully executing innovative drug discovery and development programs. Our Chief Executive Officer, Marcio Souza, previously served as Chief Operating Officer at PTC Therapeutics, Inc. and was instrumental in the development and commercialization of multiple approved products while at NPS Pharmaceuticals, Inc., Shire Human Genetic Therapies Inc. and Sanofi Genzyme Corporation. Our Chief Medical Officer, Bernard Ravina, M.D., previously Chief Medical Officer at Voyager Therapeutics, Inc., is a neurologist and movement disorder specialist who brings decades of neurologic drug development experience from roles at Biogen, the University of Rochester and the NIH’s Institute of Neurological Disorders and Stroke.

Our Approach

Each of our programs is based on four key principles that we believe will both increase the probability of success and allow us to efficiently translate insights into high-impact therapies for patients and society:

1.Focus on therapeutic targets identified through human genetics. Numerous CNS disorders are caused by an imbalance of excitation and inhibition in neuronal circuitry. By applying insights derived from the genetics of pediatric epilepsies, we have identified biological targets that we believe are implicated in determining neuronal excitability, not only in epilepsies, but also in a variety of more prevalent CNS disorders. For example, human genetics points to the relevance of the GABAergic system where mutations in GABAA receptors are associated with a number of rare pediatric epilepsies. The GABAergic system is also implicated in MDD, where enhancing GABAA activity is believed to be beneficial. As our understanding of the genetic underpinning of these disorders evolves, we plan to continually apply learnings to expand and advance our portfolio.

2.Utilize translational tools to validate the potential of our targets and product candidates. We leverage a number of translational tools to both confirm pharmacodynamic effects of our product candidates in the brain and establish on-mechanism effects, which we believe will result in an increased probability of success in the clinic. Our programs utilize target-specific EEG endpoints to serve as robust markers of pharmacological engagement of the drug target and novel transgenic animal models to assess the therapeutic activity of our molecules. We expect these tools, along with rigorous preclinical PK and pharmacodynamic characterization of our molecules will position us to more efficiently translate preclinical findings into clinical utility.

3.Pursue efficient, rigorous clinical development paths to proof-of-concept in humans. Our development strategies are focused on defining efficient paths to demonstrate the safety and therapeutic activity of our programs in humans. We select indications that we believe will enable the early demonstration of desired effect in a relatively small patient sample and we focus on clinical endpoints that both minimize inter-patient variability and offer a clear connection between pharmacodynamic effects and clinical measures that are meaningful to patients, physicians and regulatory agencies. Our global network of contract research organizations, or CROs, and scientists affords us the flexibility to conduct research and development activities in diverse geographic locations to accelerate our development timelines and limit geographic risks.

4.Apply patient-centric development strategies. We pursue the development of candidates that address the treatment needs of patients and the treating community, including targeting the underlying disease pathology versus just symptom management. We intend to develop therapies that provide patients long-term relief from their disorders and significantly reduce the overall burden to patients and caregivers. Our development strategies are tailored to demonstrate these benefits.

Our Strategy

Table of Contents

Our goal is to translate genetic insights into high-impact therapies for millions of people suffering from CNS disorders characterized by imbalance of neuronal excitation-inhibition. Key components of our strategy include:

•Advance PRAX-114 in MDD and PMD toward regulatory approval and commercialization. PRAX-114 is a potentially differentiated GABAA receptor PAM currently in Phase 2a development for the treatment of MDD and PMD. We observed marked improvements in depression scores in MDD patients within two weeks of treatment that were maintained throughout the treatment period. We are operationally ready and intend to initiate a Phase 2/3 trial in the United States and Australia by the end of March 2021. If positive, the Phase 2/3 trial in monotherapy MDD is intended to serve as one of two registrational trials required by the FDA to support clinical efficacy for the treatment of MDD, and we expect topline data in the first half of 2022. In addition to the Phase 2/3 monotherapy trial, we intend to initiate a Phase 2 DRF trial for the adjunctive treatment of MDD in the third quarter of 2021 to inform dose selection for Phase 3 monotherapy and provide controlled data to support advancing a Phase 3 adjunctive MDD trial. We are currently conducting a Phase 2a trial in Australia in PMD and expect to announce topline data in the second half of 2021. We intend to develop PRAX-114 in the United States and in other countries as both a monotherapy and adjunctive therapy for MDD and PMD for both acute and maintenance treatment. For a detailed description of the risks related to the development and commercialization of our product candidates, please refer to the section entitled “Risk Factors” in this Annual Report on Form 10-K.

•Advance PRAX-944 in ET toward regulatory approval and commercialization. PRAX-944 is a potentially differentiated selective small molecule inhibitor of T-type calcium channels in development for ET. During clinical development of PRAX-944, we have evaluated the safety and tolerability of PRAX-944 in over 150 healthy volunteers in five separate clinical trials. We have studied the safety of the PRAX-944 modified release formulation with titration up to 120mg/day and no MTD has been identified. We are currently conducting a Phase 2a proof-of-concept trial in Australia and New Zealand in ET patients. Preliminary site data from six participants of the low dose cohort showed tremor reduction, which seems to compare favorably to the standard of care agents and historical placebo response. Based on the observed safety profile and the ET participants administered up to 40mg daily, we have amended the ongoing ET trial to include a second cohort that will titrate in an open-label fashion up to 120mg daily. We have also included a randomized, double-blind, placebo-controlled withdrawal phase to the trial, where participants will either be maintained on their final open-label dose or switched to placebo. We plan to announce topline open-label safety, tolerability and efficacy data for the high dose cohort in mid-year 2021. In addition, we plan to start a Phase 1 trial to explore short titration schemes in mid-year 2021 and to initiate a Phase 2b randomized control trial in ET in late 2021. For a detailed description of the risks related to the development and commercialization of our product candidates, please refer to the section entitled “Risk Factors” in this Annual Report on Form 10-K.

•Build a rare disease franchise. We are advancing several programs for patients with rare diseases. We have six rare disease programs in our pipeline, including PRAX-562, which we believe represents the first selective persistent sodium current blocker in development for the treatment of a number of rare CNS diseases with limited or no treatment options. The current Phase 1 trial has completed the SAD portion up to maximum planned dose, with no dose limiting toxicities, and has advanced to MAD evaluation. We anticipate the initial proof of concept trial in patients to initiate in the second half of 2021. The clinical development plan for PRAX-562 encompasses exploring the broad potential for the mechanism of action in rare diseases through proof-of-concept trials in rare adult cephalgia patients and then expanding into a range of rare pediatric Developmental and Epileptic Encephalopathies, or DEEs. Given the overlapping biology, phenotypic presentation and clinical execution considerations, we believe that we can translate learnings for the treatment of DEEs across our portfolio with thorough knowledge to bring additional treatments to market. Our most advanced pre-clinical program is PRAX-222, an ASO for lowering the expression levels of the protein encoded by the gene SCN2A, in patients with GOF mutations of SCN2A epilepsy. In January 2021, the FDA granted rare pediatric disease designation for PRAX-562 for the treatment of SCN2A-DEE and SCN8A-DEE, and both rare pediatric disease and orphan drug designations for PRAX-222 for the treatment of SCN2A-DEE. Additionally, we have entered into an innovative research collaboration with The Florey Institute of Neuroscience and Mental Health to develop three novel ASOs for the treatment of rare epilepsy targets.

•Maximize the value of our product candidates through select indication expansion. All of our clinical stage product candidates address targets with therapeutic potential beyond their lead indications. As these programs advance through the clinic, we will pragmatically evaluate indication expansion and

Table of Contents

consider subsequent clinical development that will expand the labels of our product candidates to encompass other compelling opportunities at a time when we determine to be most efficient.

•Advance our understanding of genetics and neuronal imbalance to maintain our leadership and continue to build our pipeline. Advances in the field of genetics continue to elucidate new insights into mutations that drive neuronal imbalance. Our team is deeply engaged in these efforts, which we believe will enable us to pursue a pipeline discovery and development strategy grounded in these learnings and coupled with our drug discovery, translational and clinical experience. As our knowledge base continues to grow, we believe our potential to deliver additional differentiated medicines for patients will grow as well.

•Commercialize our products in the United States and globally. To realize the full potential of our product candidates, we intend to build a sales and marketing infrastructure to reach prescribers in the United States. In order to capitalize on market opportunities outside the United States, we may pursue collaborations with reputable pharmaceutical companies that have established presences in key geographies.

BROAD PSYCHIATRY AND NEUROLOGY PROGRAMS

PRAX-114

We are developing PRAX-114, an extrasynaptic GABAA receptor preferring positive allosteric modulator, or PAM, for the treatment of patients suffering from MDD and PMD. PRAX-114 is a potentially differentiated treatment for a broad MDD population, as both a monotherapy and adjunctive therapy for both acute and maintenance use. We have a multi-cohort, three-part Phase 2a clinical trial ongoing in Australia. We observed marked improvements in depression scores in MDD patients in Parts A and C of this trial within two weeks of treatment that were maintained throughout the treatment period. We expect complete topline data from Part B of the trial for the treatment of patients suffering from PMD in the second half of 2021. In October 2020, we submitted an Investigational New Drug application, or IND, to support the initiation of a Phase 2/3 clinical trial in the United States. We are operationally ready and with the recent clearance of our IND we intend to initiate a Phase 2/3 monotherapy MDD trial in the United States and Australia by the end of March 2021. If positive, the Phase 2/3 trial is intended to serve as one of two registrational trials required by the FDA to support clinical efficacy for the monotherapy treatment of MDD, and we expect topline data in the first half of 2022. In addition to the Phase 2/3 monotherapy trial, we intend to initiate a Phase 2 DRF trial for the adjunctive treatment of MDD in the third quarter of 2021 to inform dose selection for a Phase 3 monotherapy trial and to provide controlled data to support advancing a Phase 3 adjunctive MDD trial.

There is significant unmet medical need in MDD and PMD with over 22 million individuals suffering from depressive symptoms in the United States. Current pharmacological interventions suffer from multiple shortcomings including slow onset of efficacy, low remission rates and side effects that limit patient compliance. PRAX-114 targets an increasingly well-understood neuronal circuit in the brain that we believe, when properly modulated, can result in a robust and rapid antidepressant effect with an advantageous safety and tolerability profile.

We believe that our PRAX-114 program has several advantages as compared to currently available therapies and product candidates in the GABAA PAM therapeutic class:

•Planned Path to a Potential Broad MDD Label. We have been diligently pursuing our strategy to advance PRAX-114 towards regulatory approval and commercialization to support a broad label in MDD that can be easily integrated into standard clinical practice. We intend to develop PRAX-114 in the United States and in other countries as both a monotherapy and adjunctive therapy for MDD for both acute and maintenance treatment. If positive, our planned Phase 2/3 trial is intended to serve as one of two registrational trials required by the FDA to support clinical efficacy for the monotherapy treatment of MDD. In addition to the Phase 2/3 monotherapy trial, we intend to initiate a dose range finding trial for the adjunctive treatment of MDD by the end of mid- 2021 to inform dose selection for Phase 3 and provide controlled data to support advancing a Phase 3 adjunctive MDD trial.

•Wider Therapeutic Window. We have determined that PRAX-114 is approximately 10-fold more selective PAM of the extrasynaptic form of GABAA receptors compared to the synaptic form. In healthy volunteers, we have observed PRAX-114 to markedly increase quantitative electroencephalography, or qEEG, power in the alpha and beta-frequency bands—unlike GABAA receptor PAMs that only modulate synaptic GABAA receptors, such as benzodiazepines, or that are equipotent at synaptic and extrasynaptic

Table of Contents

receptors, such as allopregnanolone, which decreases power in the alpha frequency band. We believe these data suggest that PRAX-114 has a differentiated pharmacological profile to other GABAA PAMs at therapeutic doses due to the relatively selective activation of extrasynaptic GABAA receptors. By preferentially modulating extrasynaptic GABAA receptors, we believe PRAX-114 is able to uniquely activate the GABAergic target and has the potential to mediate antidepressant and anxiolytic activity without the significant sedation observed with less selective neuroactive steroids.

•Simple Nightly Dosing. We believe the ability to administer PRAX-114 and achieve targeted exposures, with or without food, is key for clinical and commercial success in MDD. This is also critical for a patient-centric therapy because many patients with depression suffer from appetite disturbances. We have observed fast absorption of PRAX-114 within one to three hours of dosing and a predictable PK profile across multiple trials. Based on clinical findings to date, PRAX-114 achieves reproducible overall exposure (i.e., area under the concentration curve, or AUC) across a wide range of administration conditions, demonstrating consistent exposure when administered with or without food and at different times of day, whereas other GABAA PAM neuroactive steroids may require food to achieve therapeutic levels. While AUC is unaffected by administration conditions, nightly dosing has been demonstrated to reduce Cmax thereby enhancing the potential for improved tolerability.

•Sustained Administration. After consultation with the FDA and other stakeholders in MDD and PMD therapy, we designed our Phase 2/3 trial of PRAX-114 to include 28-day nightly dosing to evaluate patients at 14 days to assess the rapidity and robustness of response and 28 days to measure initial of effect. We believe that having a dosing paradigm consistent with the duration of depressive episodes and easily integrated into standard clinical practice will provide the most substantial benefit to patients in controlling their disease, further differentiating PRAX-114 from other GABAA PAMs.

•Indication Expansion. Based on the novel pharmacology of PRAX-114 and its generally well-tolerated profile in clinical trials to date and our knowledge of disorders related to MDD that may be treatable through the GABAA PAM mechanism, we believe PRAX-114 is suitable for potential development across a wide-range of indications in psychiatry and neurology, providing for potentially sizable expansion opportunities to explore in addition to MDD.

Major Depressive Disorder

Major Depressive Disorder, or MDD, is a chronic psychiatric condition causing severe impairments that interfere with the ability to carry out life activities. An MDD episode is characterized by a period of at least two weeks of persistent depressed mood and/or the loss of interest or pleasure in activities, accompanied by sleep and appetite disturbance, fatigue, concentration difficulty, cognitive impairment, feelings of guilt, psychomotor retardation or agitation and suicidal ideation. MDD is one of the most prevalent psychiatric disorders. In the United States, approximately 19 million adults, or 7% of the adult population suffer from MDD, with episodes lasting on average six to eight months. It is estimated that MDD affects more than 300 million people worldwide. Moreover, the prevalence of depression has increased during the COVID-19 pandemic in the US and globally. In the United States, depression symptoms have increased by more than 3-fold overall during the COVID-19 pandemic. The most dramatic increases are reported in moderate, moderately severe and severe depression symptoms, with a 2.6-fold, 3.7-fold, and 7.5-fold rise, respectively, relative to a pre-COVID-19 pandemic period.

MDD is a recurrent psychiatric condition that frequently requires long-term treatment, with the ultimate goal of achieving remission. MDD is associated with an elevated risk of suicide, underscoring the need for rapid and effective treatment. The most explored pharmacological mechanisms for treating MDD target monoamine neurotransmitters. Drugs in this class include selective serotonin reuptake inhibitors, or SSRIs, serotonin and norepinephrine reuptake inhibitors, or SNRIs, bupropion and atypical antipsychotic medications. SSRIs and SNRIs are associated with significant side effects, including weight gain, sexual dysfunction, drowsiness, nausea, insomnia and discontinuation syndrome. Atypical antipsychotics indicated for adjunctive treatment of insufficient clinical response are associated with weight gain, sexual dysfunction, metabolic syndrome and movement disorders. The side effect profile of current antidepressant standard of care negatively impacts treatment outcomes, quality of life and adherence in MDD patients.

Approximately seventy percent of MDD patients fail to achieve remission with first line treatment. Further, those patients that are responsive typically require approximately six to eight weeks of treatment to show a clinically meaningful response. Slow onset of action is a substantial unmet need in MDD, with some of the most commonly prescribed antidepressants showing a reduction in the Hamilton Depression Scale, or HAM-D, of approximately 6- to 8-points and a difference from placebo of approximately 1-2 points at Week 2 (Figure 1). Moreover, approximately

Table of Contents

40% of patients on therapy discontinue treatment due to either a loss of response or adverse side effects. Finally, 33% of patients fail to respond after treatment with three or more different standard of care therapies.

Among the MDD patients who experience a response to treatment, the majority do not achieve remission. Even for patients deemed responsive, disease burden often persists through the presence of residual depression symptoms that lead to an ongoing negative impact on home, interpersonal and occupational functioning, as well as a significantly increased risk of relapse of the full depressive syndrome and worse comorbid outcomes, including suicide.

Despite the numerous and long-standing antidepressant treatment options, there continues to be an unmet need for antidepressants that provide rapid onset of effect, higher remission rates, efficacy throughout the depressive episode and an improved tolerability profile that is aligned with the clinical care and the course of MDD and its accompanying comorbid symptoms.

Figure 1. Reduction of HAM-D from baseline of commonly prescribed antidepressants and placebo at Week 2.

Perimenopausal depression

Perimenopause is the transition between the onset of hormonal and clinical features of menopause and the one-year period after the final menses. Perimenopause can last up to 10 years. Women with no lifetime history of major depression who have entered the perimenopause period are found to be twice as likely to develop significant depressive symptoms as women who have not entered the perimenopause period. Notably, the increased risk for depression during the perimenopausal transition has been observed to decline substantially after the final menstrual period.

There are over 30 million women in the United States between the ages of 45 and 59 years who are at risk of developing perimenopause symptoms, with an estimated three million developing mood symptoms such as depression, anxiety, irritability and suicidal ideation and behavior and an estimated 20 million women developing associated vasomotor symptoms or hot flushes. Notably, suicide rates are the highest among women 45 to 59 years of age and have increased by approximately 42% in recent decades.

Although primarily viewed as a reproductive transition, the symptoms of perimenopause are largely neuropsychiatric in nature. Neurological symptoms that emerge during perimenopause are indicative of disruption in multiple estrogen and progesterone-regulated systems such as thermoregulation, sleep, circadian rhythms and sensory processing and affect multiple domains of cognitive function. Perimenopausal depression also appears to impact the clinical symptomatology of menopause, with the presence of depression being associated with a greater degree of menopausal hot flushes than in women without perimenopausal depression.

There is substantial evidence that fluctuations in estrogen and progesterone, the precursor of the endogenous neuroactive steroid allopregnanolone, a GABAA receptor PAM, are in part responsible for the mood changes, hot flushes and other neurologic symptoms of perimenopause. Similar to MDD, SSRIs and SNRIs have shown limited efficacy in treating perimenopausal depression. There remains an unmet medical need for effective treatment of core depression symptoms and associated physical symptoms of menopause.

Table of Contents

A GABAA receptor PAM, like PRAX-114, that potentiates the activity of endogenous neuroactive steroids on GABAA receptors, may offer broader therapeutic benefit compared to current standard of care antidepressants.

GABAA in depression

Gamma-aminobutyric acid, or GABA, is the principal neurotransmitter mediating neuronal inhibition in the brain. Neurons that produce GABA, known as GABAergic neurons, are present throughout the brain, representing between 20 percent and 40 percent of all neurons depending on the region. Their primary role is to balance and fine tune excitatory neurotransmission of various neuronal circuits. Whole-exome sequencing has identified GABAA receptor mutations as an important cause in a range of neurological conditions, underscoring their importance as central regulators of excitatory and inhibitory balance in the brain.

It is well established that GABAergic signaling is impaired in patients with MDD and other stress-related mood disorders. GABA levels, levels of the GABA synthesizing enzyme GAD67, as well as GABAA receptor levels, have been shown to be reduced in brains of patients with MDD. In addition, decreased GABAergic neuron function, most notably in the prefrontal cortex, has been documented in MDD patients and in preclinical animal models of depression. Endogenous neuroactive steroids, such as allopregnanolone and pregnanolone or synthetic derivatives thereof, such as PRAX-114, are known to potentiate the activity of GABAA receptors. Both human and animal data reveal an important role for neuroactive steroids in these GABAergic deficits and levels of endogenous neuroactive steroids are decreased in individuals with MDD and PMD.

Of particular relevance to the PRAX-114 program is the more recently established link between GABAergic signaling, neuroactive steroid levels and stress—a well-established risk factor for MDD and other mood disorders. In preclinical models, exposure to chronic stress leads to reduced neuroactive steroid biosynthesis and reduced GABAergic inhibition in depression-relevant brain circuits. This ultimately results in increased anxiety and depression-like behaviors. In particular, it has been shown that stress causes long-lasting loss of GABAergic inhibition in the amygdala, a brain region central to the stress response involved in controlling emotions. This reduced inhibition causes increased activity of the amygdala and is associated with an exaggerated stress hormone response.

We believe that enhancing modulation of GABAA receptors in patients with depression and anxiety has the potential to restore normal function in these circuits, leading to broad applications in mood and anxiety disorders.

GABAA receptors: The target of PRAX-114

PRAX-114 is a small molecule neuroactive steroid that acts as a positive allosteric modulator of GABAA receptors. Positive allosteric modulators, or PAMs, are substances that bind to a receptor, such as GABAA, to enhance that receptor’s response to its endogenous ligand (or endogenous agonist). GABAA PAMs bind to a distinct site from endogenous GABA, an allosteric binding site, and do not activate the receptor in the absence of the GABA. Allosteric modulators are believed to have improved safety profiles and are less likely to result in tachyphylaxis, or decreasing drug response, as compared to agonists. GABA exerts its effects through binding to two types of GABAA receptors, synaptic and extrasynaptic receptors, which differ in their protein subunit composition, physical location on the cell surface and functional role in modulating neuronal circuits.

GABAA receptors are composed of five subunits which include two alpha, two beta and a fifth subunit (either gamma or delta) that is dependent on the type of receptor. Synaptic GABAA receptors, which are located in the synapse of neurons, contain a gamma subunit while GABAA receptors located outside of the synapse, referred to as extrasynaptic GABAA receptors, contain a delta subunit. Molecules that act as PAMs of only the synaptic GABAA receptor, such as benzodiazepines, bind to sites situated at the interface between the alpha and gamma subunits. Molecules that act as PAMs of both synaptic and extrasynaptic GABAA receptors, such as the neuroactive steroids allopregnanolone and PRAX-114, bind to sites situated at the interface between the alpha and beta subunits present in both types of receptors. The figure below displays the synaptic binding site for drugs such as benzodiazepines, and the distinct extrasynaptic and synaptic binding sites for neuroactive steroids, such as allopregnanolone and PRAX-114.

Table of Contents

Figure 2. GABAA synaptic and extrasynaptic receptors differ in structure and function.

Synaptic and extrasynaptic GABAA receptors have distinct functions. Synaptic GABAA receptors are responsible for short-lasting, or phasic, inhibition of neurons in response to GABA release at the synapse. By contrast, extrasynaptic GABAA receptors drive continuous, or tonic, low-level inhibition of neurons in response to lower ambient levels of GABA outside of the synapse. While synaptic receptors can respond quickly to stimulation and network demand, extrasynaptic receptors have a broader modulatory role, serving to continuously modulate the overall excitability of neurons.

Molecules that act as PAMs of only the synaptic GABAA receptor, such as benzodiazepines and barbiturates, are used for sedation, sleep induction and anxiolysis, and have anticonvulsant and muscle relaxant properties. These drugs have potent and rapid onset of activity but have not demonstrated antidepressant effects.

Allopregnanolone is an endogenous neuroactive steroid and a PAM of both the extrasynaptic and synaptic GABAA receptors, which has been associated with antidepressant activity. However, allopregnanolone also has shown significant dose-limiting sedative activity, which we believe is likely mediated at least partially by its effects on synaptic GABAA receptors. Despite this limitation, a formulation of allopregnanolone has been approved and is marketed as ZulressoTM to treat post-partum depression.

The distinct effects mediated by these classes of GABAA PAMs suggest that modulation of extrasynaptic GABAA receptors is responsible for the antidepressant effects demonstrated by allopregnanolone. One of the goals for a next generation neuroactive steroid, such as PRAX-114, is to preferentially modulate extrasynaptic GABAA receptors while minimizing the sedative impact from modulation of synaptic GABAA receptors.

PRAX-114 preference for extrasynaptic GABAA receptors

To assess the relative potency in-vitro of PRAX-114-mediated GABAA receptor activation for synaptic and extrasynaptic receptors, we measured the peak current induced by a low concentration of GABA (2 μM) in the presence of increasing concentrations of PRAX-114 in Chinese Hamster Ovary, or CHO, cells expressing either extrasynaptic (α4ß3δ) or synaptic (α1ß2γ2) human GABAA receptors. In this model, PRAX-114 potentiates the GABA-activated current of both extrasynaptic and synaptic GABAA receptors, but was approximately 6.4-fold more potent in potentiating the extrasynaptic form of the receptor than the synaptic form based on the concentration that gives half-maximal response, or EC50. At a concentration that activates extrasynaptic GABAA receptors to the equivalent of full activation by the endogenous ligand GABA (~260 nM, 300% potentiation of 2μM GABA), PRAX-114 led to 10.5-fold greater potentiation of extrasynaptic GABAA receptors than synaptic GABAA receptors (29%) (Figure 3, Table 4).

Table of Contents

Figure 3. At 300% extrasynaptic GABAA receptor potentiation (equivalent to ~100% activation by endogenous agonist GABA), PRAX-114 led to 10.5-fold greater potentiation of extrasynaptic GABAA receptors than synaptic GABAA receptors (29%).

In the same assay, at the same level of extrasynaptic GABAA receptor potentiation (300%), other GABAA receptor PAM neuroactive steroids in development, or on the market, demonstrated only 0.4 to 2.6 fold greater potentiation of extrasynaptic GABAA receptors, which compares unfavorably to the 10.5 fold observed for PRAX-114 (Table 4). Based on these assay conditions, we believe that the differentiated preference at extrasynaptic GABAA receptors by PRAX-114 will allow it to achieve high levels of extrasynaptic GABAergic activation with improved tolerability.

ZulressoTM (brexanolone) IV 300 % 306 % 1.0

Table 4. Comparison of the degree ofin-vitroGABAA receptor potentiation achieved by PRAX-114 and other neuroactive steroid GABAAPAMs.α4β3δ: extrasynaptic GABAA receptors,α1β2γ2: synaptic GABAA receptors.

PRAX-114 clinical development in depression

We have initiated clinical development for PRAX-114 in mood disorders. To date, two Phase 1 clinical trials of PRAX-114 have been completed in healthy volunteers. These studies in 82 healthy volunteers showed PRAX-114 to be generally well-tolerated, with dose-dependent pharmacodynamic activities. In our ongoing Phase 2a clinical trial in Australia, we observed marked improvements in depression scores in MDD patients within two weeks of treatment that was maintained throughout the treatment period. We have also conducted a pharmacokinetics bridging study in healthy volunteers that confirmed our generally well-tolerated profile and consistency in exposure across a wide range of administration conditions including fed versus fasted and morning versus evening dosing. We are operationally ready and intend to initiate a Phase 2/3 trial in the United States and Australia by the end of March 2021. If positive, the Phase 2/3 trial is intended to serve as one of two registrational trials required by the FDA to support clinical efficacy for monotherapy treatment of MDD, and we expect topline data in the first half of 2022. In addition to the Phase 2/3 monotherapy trial, we intend to initiate a Phase 2 DRF trial for adjunctive treatment of MDD by mid-2021 to inform dose selection for the Phase 3 monotherapy trial and provide controlled data to support advancing a Phase 3 adjunctive MDD trial.

Table of Contents

Phase 1 SAD and MAD trials in healthy volunteers

We conducted a Phase 1 randomized, double-blind, placebo-controlled single ascending dose, or SAD, trial of PRAX-114 in healthy volunteers to evaluate safety and tolerability of PRAX-114. This trial enrolled 36 volunteers who were randomized into cohorts dosed with 1mg, 3mg, 10mg, 30mg or 60mg of PRAX-114 or placebo. PRAX-114 was generally well-tolerated and no serious adverse events, or SAEs, were reported in this trial.

We subsequently conducted a Phase 1 randomized, double-blind, placebo-controlled MAD trial in healthy volunteers in Australia to evaluate the safety, tolerability and pharmacokinetics of PRAX-114 and to assess the effect of food on drug exposure. Thirty-six volunteers were randomized to receive daily doses of 15mg, 30mg or 60mg of PRAX-114 or placebo for 14 days. Ten additional volunteers in a food effect cohort received 30mg doses of PRAX-114 when they were in a fasted state or with a high-fat meal.

As part of our MAD trial, we measured the effect of PRAX-114 on the quantitative EEG, or qEEG, to understand the pharmacodynamic effect of PRAX-114 on GABAA receptor activation. An EEG is a real-time non-invasive measure of electrical activity of neurons in the brain. The frequency and amplitude of the detected electrical signals provide insights into brain function and brain state (e.g., awake, deep sleep, etc). qEEG, also called pharmaco-EEG, is a quantitative measure of the changes in brain activity in specific EEG frequency bands in response to treatment with a brain-active compound. Changes in power in the beta frequency band, specifically, are used as a pharmacodynamic biomarker of GABAA receptor activation in response to a brain active compound.

In both the Phase 1 SAD and MAD trials, we observed fast absorption of PRAX-114 within one to three hours of dosing and approximately dose-proportional increases in peak concentration and total drug exposure. In the MAD trial, the half-life of the drug was between 12.2 and 14.8 hours, consistent with a once-daily dosing paradigm. Little or no accumulation of the drug was observed in the multiple dose trial over the ranges of doses tested.

We believe that the simple nightly administration of PRAX-114 with or without food is key for clinical and commercial success in MDD, as many patients struggle with adherence to medication and forcing a dietary regimen would impose further complications in this vulnerable population. In the food effect cohort of the MAD trial, overall drug exposure as measured by area under the concentration curve, or AUC, of PRAX-114 increased by only 1.17-fold in the fed state versus in the fasted state. The primary effect of food was observed in the Cmax, which was 0.64-fold of that observed under fed conditions. These findings indicate that PRAX-114 does not need to be taken with food to achieve therapeutic exposures. According to results presented in a published patent application, zuranolone, a GABAA PAM neuroactive steroid in development for treatment of MDD, exhibited a food effect that resulted in increases in Cmax of approximately 2.88-fold and in AUC of approximately 1.58-fold in the fed versus the fasted conditions (Table 5). We believe this food effect has led to the development of zuranolone requiring administration with a high-fat meal for the compound reliably achieve target exposures. We believe that the absence of a requirement that PRAX-114 be taken with food creates a potential competitive advantage over drugs that may require administration with food to achieve consistent target exposures, and should allow flexibility to adjust to the comorbid changes in appetite and preferences of MDD patients.

Table 5. Food effect clinical studies of PRAX-114 and Zuranolone (WO2019/051264 A1). Subjects were administered a high-fat meal 30 minutes prior to administration of the compound in both studies.

Table of Contents

In the MAD trial, PRAX-114 was generally well-tolerated, with no SAEs reported. The reported treatment-emergent adverse events, or TEAEs, were mild to moderate and were consistent with those expected for the mechanism of action. The most common adverse event was somnolence, which is characterized by sleepiness or drowsiness, and was reported by 78% (7/9) of those subjects receiving the 60mg dose; all events of somnolence were mild in severity. Increases in sleepiness as measured by the Stanford Sleepiness Scale occurred between one- and three-hours post-dosing, consistent with the period of peak drug concentrations; sleepiness ratings at the 60mg dose were similar to placebo within 4 hours post-dose. We did not observe a maximally tolerated dose, or MTD.

In our Phase 1 multiple ascending dose trial of PRAX-114 in healthy volunteers in Australia, we also observed the following TEAEs, all of which were mild to moderate in severity:

•60mg dose (n=9), we observed nervous system disorder TEAEs, of somnolence (77.8% of subjects), headache (33.3% of subjects), dizziness (55.6% of subjects) and hypoaesthesia, or diminished sense of touch (22.2% of subjects). Other TEAEs observed in more than one subject were euphoric mood (22.2% of subjects), hyperhidrosis, or excessive sweating (22.2% of subjects) and muscle twitching (22.2% of subjects).

•30mg dose (n=9), we observed nervous system disorder TEAEs of somnolence (44.4% of subjects) and headache (22.2% of subjects). Other TEAEs observed in more than one subject were skin irritation (55.6% of subjects) and euphoric mood (22.2% of subjects).

•15mg dose (n=9), the only TEAE observed in more than one subject was fatigue (22.2% of subjects).

•Placebo group (n=9), we observed fatigue (22.2% of subjects).

TEAEs appearing to be dose related were somnolence, dizziness, headache, euphoric mood and hypoaesthesia.

We measured changes in qEEG power in our Phase 1 MAD volunteers to assess the effect of PRAX-114 on GABAA receptors in the brain on days 1 and 14 of this trial. PRAX-114 produced marked increases in the power of the alpha and beta-frequency bands. Increases in the beta-frequency band are correlated with GABAergic activation, as previously shown by the marketed GABAA PAMs, allopregnanolone and lorazepam. PRAX-114 distinctly increases the alpha-frequency band, unlike allopregnanolone and lorazepam, which have shown to decrease the power of the alpha-frequency (Figure 6). We believe this qEEG profile of PRAX-114 is consistent with its extrasynaptic GABAA receptor preference and differentiated pharmacological profile relative to benzodiazepines and other GABAA PAMs in the class.

Figure 6. GABAA PAM qEEG signal of PRAX-114, allopregnanolone, and lorazepam. (Adapted from 1. SAGE R&D Day Presentation, 2. Gilles et al. 2002)

Moreover, the PRAX-114 increases in alpha and beta-frequency were strongly correlated with dose and PRAX-114 levels in the blood. In these healthy volunteers at one hour post-dose on Day 1, PRAX-114 30mg resulted in an average increase in qEEG alpha and beta power of approximately 1.5-fold and 1.6-fold compared to baseline, and 60mg resulted in an increase in this measure of 2.6-fold and 2.8-fold compared to baseline, respectively (Figure 7). The effects on the qEEG alpha and beta power were sustained at Day 14. These data show that PRAX-114 engaged GABAA receptors in the brain and produced consistent effects on qEEG within the first hour after dosing with similar effects on Days 1 and 14. This finding also supports comparison and translation of the

Table of Contents

pharmacologic activity and qEEG data from the pre-clinical studies, where a 1.6-fold increase in beta power was associated with robust activity in animal models of anxiety and depression and was used to inform dose selection in subsequent clinical trials.

Figure 7. PRAX-114 (30mg and 60mg) showed a robust dose-dependent qEEG signal and target activation that was sustained over 14 days of dosing.

Notably, in the MAD study, PRAX-114 showed increases in beta power up to 2.8-fold without achieving a MTD or demonstrating any SAEs, and a higher dose of 80mg dosed as a single dose in healthy participants for 14 days was also generally well-tolerated in a subsequent clinical trial in MDD, summarized below. In a separate study, another molecule in development in the class resulted in degrees of sedation that were not tolerated at doses that resulted in increases in beta power by approximately 1.7-fold compared to baseline. We believe this highlights the extrasynaptic GABAA receptor preference and unique pharmacological profile of PRAX-114 and its ability to achieve high levels of GABAergic activation with improved tolerability.

This improved tolerability profile of PRAX-114 offers the potential for a wider therapeutic window, increased adherence and a wider dose range for MDD patients.

Phase 2a trial in patients with depression

Based on the observed pharmacology in the Phase 1 trials, we are currently conducting a three-part, open-label, Phase 2a trial in Australia to assess the safety and efficacy of PRAX-114 in patients with moderate to severe MDD or PMD. We have completed Parts A and C of this Phase 2a trial and Part B is ongoing.

Part A results

Part A of the open-label trial included two weeks of treatment and was designed to evaluate the timing and magnitude of the antidepressant effects of PRAX-114 across a range of doses in patients with MDD. Patients were required to be between the ages of 18 and 65 and to have moderate to severe MDD for at least one month as defined by the Hamilton Depression Rating Scale, or HAM-D, score of 22 or higher. The HAM-D, one of the most widely-used clinical rating scales for depression, was the main assessment used to quantify levels of depression in these patients. The 17 items used for scoring this scale cover a wide range of symptoms typically found with depression including mood, suicidal thoughts, insomnia, anxiety, loss of appetite and weight loss. Patients with more severe depression have higher scores. The effect of PRAX-114 was measured by the change in the HAM-D score relative to baseline. Patients who had previously failed to respond to a standard of care antidepressant in their current episode were eligible for inclusion. In addition to HAM-D, other scales used included the Montgomery–Åsberg Depression Rating Scale, or MADRS, the Hamilton Anxiety Rating Scale, or HAM-A, and the Symptoms of

Table of Contents

Depression Questionnaire, or SDQ. MADRS is a 10 item rating scale designed to assess the severity of symptoms in a depressive illness. HAM-A is a 14 item scale widely used to measure the severity of anxiety symptoms, including both psychic anxiety (mental agitation and psychological distress) and somatic anxiety (physical complaints related to anxiety). SDQ is a 44 item self-reported scale designed to measure the severity of symptoms across several subtypes of depression, including irritability, anger attacks and anxiety.

We selected an initial target dose of 45mg daily of PRAX-114 suspension formulation that was expected to achieve exposures demonstrating full clinical improvement based on the Phase 1 data and qEEG findings. Two additional cohorts were subsequently conducted to assess higher daily doses of 60mg and 80mg PRAX-114 due to the generally well-tolerated profile at 45mg. The first week of treatment was conducted in an inpatient setting to facilitate daily efficacy and safety assessments and then patients were discharged and treated as outpatients for the second week. Patients were instructed to take PRAX-114 at 4:00 PM on Day 1 to support collection of post-dose pharmacokinetics samples and then nightly at 9:30 PM on Days 2-14. Patients were not required to take PRAX-114 with food. Compliance was carefully monitored throughout the duration of the trial, including inpatient and outpatient periods, with the AiCure smartphone adherence monitoring system.

Thirty-three patients were enrolled and completed Part A before the COVID-19 pandemic began impacting clinical trial conduct globally. At baseline, patients had a mean HAM-D total score of 25, ranging from 20 to 33, indicating moderate to severe MDD. Twenty-six of the thirty-three participants had previously received an antidepressant during the current depressive episode but still had moderate to severe MDD. This failure to respond to initial antidepressant treatment has been associated with more severe and refractory depression. The remaining patients were not being treated with any antidepressant for the current episode before enrolling in the trial.

Dosing with PRAX-114 led to a marked improvement in the HAM-D score (Figure 8) within two weeks of treatment. After one week of treatment, least squares, or LS, mean improvements of 15 to 19 points from baseline were noted across the three dose groups. After two weeks of treatment, all 3 dose levels showed improvements from baseline of greater than 13 points with mean improvements from baseline of 14 to 16 points. Across all dose levels, two-thirds of patients were responders (defined as a >=50% reduction in HAM-D) or were clinically in remission (HAM-D<=7) at the end of the 14 day treatment period. Changes in other depression-related scales measured such as MADRS, HAM-A and SDQ were consistent with the changes in HAM-D. While the study was not powered to show differences between dose levels, there was no notable dose response observed, which is common amongst trials of antidepressants.

Figure 8. Reduction in HAM-D total score observed in MDD patients treated with PRAX-114.

After 14 days of treatment, patients were monitored for an additional 14 days. During this monitoring period, the core mood symptoms measured by the HAM-D generally remained stable with a slight increase in the insomnia item scores post-treatment.

Table of Contents

While our Phase 2a trial is not placebo controlled, prior randomized placebo-controlled MDD trials provide important context for the interpretation of the clinical response. The marked improvements in HAM-D scores seen in MDD patients in Part A within two weeks of treatment compares favorably to published reports on changes in HAM-D scale in clinical trials of approved antidepressants such as vortioxetine and duloxetine, among others, which commonly take approximately six to eight weeks to reach a maximal efficacy and often fail to differentiate from placebo at two weeks. Moreover, mean HAM-D changes from baseline at Day 14 for the placebo group of these randomized controlled antidepressant trials are most often between 4-8 points. Even at the first post-dose assessment on Day 3, patients dosed with PRAX-114 had a mean decrease of over 11 points on the HAM-D scale, which compares favorably with the average changes reported in the placebo groups at Day 14 from randomized studies completed for recently approved antidepressants (Figure 9), and other common antidepressants after several weeks of dosing. The clinical data that we have generated to date, and that we expect to generate in the future, from our clinical studies will constitute the bulk of the data needed to support an application for marketing approval of PRAX-114. Unless we conduct head-to-head studies of PRAX-114 against other molecules as part of our future clinical trials and elect to include the resulting data in an application for regulatory approval, we would not expect to rely upon PRAX-114’s potential differentiation from any other molecules in connection with submissions to the FDA or other regulatory agencies, as applicable, for approval or otherwise. As the data presented above is based on a cross-trial comparison and not a head-to-head clinical trial, such data may not be directly comparable due to differences in study protocols, conditions and patient populations. Accordingly, cross-trial comparisons may not be reliable predictors of the relative efficacy or other benefits of PRAX-114 compared to other product candidates that may be approved or that are or were in development for MDD.

Figure 9. Change from baseline to Day 14 in HAM-D total score for the Placebo arm of selected randomized placebo-controlled studies of recently approved antidepressants. Bubble size is proportional to the sample size of the placebo group in each study. Across studies, the sample size in the placebo group ranged from 89 to 277.

PRAX-114 was generally well-tolerated across the dose range, including at the highest 80mg dose. This is consistent with an expected wider therapeutic window based on the preferential selectivity of PRAX-114 for extrasynaptic GABAA receptors. TEAEs were generally mild to moderate. Rates of somnolence, which is characterized by sleepiness or drowsiness, increased with dose, demonstrating a pharmacological effect which was somewhat mitigated by dosing at night versus the morning. With night-time dosing, 12/33 patients (36%) noted somnolence post-dosing, which was generally time-limited, not experienced during the daytime and substantially lower than the 78% reported in the Phase 1 60mg group with morning dosing. There were no SAEs or discontinuations and study drug cessation at the end of the treatment period was generally well-tolerated.

In this MDD part of the trial, we also observed the following TEAEs in at least 2 subjects per dose level:

Table of Contents

•80mg dose (n=7), headache (42.9% of subjects), somnolence (42.9% of subjects), dizziness (57.1% of subjects), feeling drunk (28.6% of subjects) and diarrhea (28.6% of subjects).

•60mg dose (n=13), headache (46.2% of subjects), somnolence (53.8% of subjects), dizziness (30.8% of subjects), feeling drunk (23.1% of subjects) and constipation (23.1% of subjects).

•45mg dose (n=13), headache (53.8% of subjects), somnolence (15.4% of subjects), fatigue (23.1% of subjects), vessel puncture site bruise from blood draws (15.4% of subjects), abdominal distension (15.4% of subjects) and upper respiratory tract infection (15.4% of subjects).

Across all dose levels studied, headache (48.5%), somnolence (36.4%), dizziness (24.2%), fatigue (15.2%), feeling drunk (15.2%), constipation (12.1%) and vessel site puncture bruise (12.1%) were reported in >10% of subjects.

Part B ongoing

We are currently conducting Part B of this trial in order to assess PRAX-114 in PMD patients. We are dosing up to twelve patients with PMD with 60mg of PRAX-114 nightly at 9:30PM for 14 days on an outpatient basis. The dose for this part of the trial was selected based on the data from Part A. Inclusion criteria for Part B are similar to Part A and C, except that it requires participants to be females of 40 years of age or older with irregular menses and hot flushes. Part B will help to determine if PRAX-114 has an effect on broader menopausal symptoms, like hot flushes, in addition to confirming the antidepressant effect. We anticipate topline results in the second half of 2021.

Part C results

Part C of this trial has been completed. The goal of Part C was to evaluate the safety of four-week outpatient dosing with PRAX-114, similar to the study duration of our Phase 2/3 trial, and the treatment effect profile from Day 15 to Day 28.

Inclusion criteria and symptom assessments were the same as Part A. Part C, however, was conducted in Melbourne Australia from June through October 2020, a period of highly restrictive public health lockdown due to the COVID-19 pandemic with closing of non-essential businesses, limited access to healthcare facilities, nightly curfews and restriction of residents to their homes except for essential healthcare or safety reasons. Only one person per household could travel up to 5 km to purchase essential supplies once per day. While the conduct of Part C continued through this period, the public health intervention required an abrupt change to the use of telehealth administered clinical efficacy assessments, mailed self-report assessments, and courier delivery of study drug to participants. These revisions in study conduct successfully supported consistent site and participant adherence to study procedures and study drug administration through completion of Part C. The experience managing these impacts has been integrated into the design and operationalization of the planned Phase 2/3 clinical trials with PRAX-114 in MDD.

A total of thirteen participants were enrolled and completed a nightly 9:30 PM dose of PRAX-114 at 60mg for four weeks. At baseline, patients had a mean HAM-D total score of 25, that ranged from 22 to 30, indicating moderate to severe MDD. Eight of the thirteen participants had previously received an antidepressant during the current depressive episode but still had moderate to severe MDD. This failure to respond to initial antidepressant treatment has been associated with more severe and refractory depression. The remaining patients were not being treated with any antidepressant for the current episode before enrolling in the trial.

PRAX-114 was generally well-tolerated, without a change in safety profile, after four-weeks of 60mg once nightly 9:30 PM outpatient dosing in Part C. We observed no new patterns of AEs in the Day 15 to Day 28 treatment period or post-discontinuation after four-week treatment compared to Part A.

In this part of the MDD trial, we observed the following TEAEs in at least 2 subjects with 60mg PRAX-114:

•Headache (46% of subjects), somnolence (31% of subjects), feeling abnormal (15% of subjects), fatigue (15% of subjects), dry mouth (15% of subjects), nasopharyngitis (15% of subjects), weight decreased (15% of subjects) and nausea (38% of subjects).

•Examination of the nausea AEs, which have been infrequent in prior studies, found patterns inconsistent with direct pharmacological effects in four out of five cases (two cases with onset immediately upon drinking PRAX-114 suspension, one case with 10 min duration in the morning for 7 days, and one case preceding an AE of nasopharyngitis).

Table of Contents

Dosing with PRAX-114 for four weeks led to a rapid and marked improvement in the HAM-D score (Figure 10) within two weeks of treatment, with LS Mean improvement of 11 points at Day 15 remained stable through the end of the active treatment period).

Figure 10. Reduction of HAM-D total score observed in MDD patients treated with PRAX-114 in Part C.

Data from Part B to date shows a safety profile similar to Parts A and C. The preliminary data show that dosing with PRAX-114 led to a marked improvement in the HAM-D score within two weeks of treatment. We expect complete topline data from Part B of the trial in the second half of 2021.

More than 70% of participants from all parts of the study had previously received an antidepressant during the current depressive episode but still had moderate to severe MDD. Preliminary data show greater than 60% of patients were responders or clinically in remission at two weeks. All cohorts demonstrated good tolerability.

Pharmacokinetics bridging study

To date, we have conducted all studies with PRAX-114 using a liquid, suspension formulation. For our registrational studies and subsequent potential commercial use of PRAX-114, we have developed a tablet formulation.

In a clinical PK bridging study, PRAX-114 was administered as a solid dose (i.e., tablet) formulation in single ascending doses of 40mg, 60mg and 80mg and compared to PRAX-114 suspension administered at the 60mg dose. The PK of the tablet formulation was found to be comparable to the suspension formulation at the 60mg dose level (Figure 11). Administration of the 60mg tablet formulation under fasted conditions resulted in a median tmax of ~1.0 hour, Cmax of ~400ng/mL, AUCinf of ~2600 hr*ng/mL and t1/2 of ~11-12 hours, similar to the 60mg oral suspension. Exposure to PRAX-114 increased approximately proportional to dose.

Table of Contents

Figure 11. PRAX-114 dose-ranging pharmacokinetic study of tablet formulation and PRAX-114 pharmacokinetic bridging study of suspension and tablet formulations. Arithmetic means are displayed with standard deviations at each collection time point.

When PRAX-114 is administered prior to bedtime on an empty stomach or with a high-fat or high- calorie meal, the rate of absorption is reduced, resulting in a delay in tmax and reduction in Cmax, while AUC is generally unchanged as compared to fasting conditions. Thus, PRAX-114 maintains consistent overall exposure across a wide range of administration conditions within the evaluated dose range, and the reduction in Cmax observed with evening dosing appears to be associated with an improved tolerability profile. Daily administration of 20, 40 and 60mg in the morning under fasted conditions for 14 days resulted in increases in exposure (Cmax and AUC0-24) from Day 1 to Day 7 and was generally comparable between Day 7 and Day 14, with Cmax accumulation ratios of approximately 1.3 consistent with the half-life of PRAX-114 and anticipated achievement of steady-state within two to three days.

Overall, the PK profile of PRAX 114 has demonstrated consistent drug exposure across a wide range of administration conditions, demonstrating a consistent AUC when administered with or without food and at different times of day, and enabling once daily administration at bedtime without the need for additional patient instructions.

Planned PRAX-114 clinical trials

In October 2020, we submitted an IND to support the initiation of a Phase 2/3 clinical trial in the United States and Australia in approximately 200 moderate to severe MDD patients. At the end of the 30-day review period, the FDA notified us that the IND was placed on full clinical hold pending the resolution of certain non-clinical pharmacology and toxicology matters. We subsequently interacted with FDA to gain agreement on a path to initiate the clinical study, which included a proposal to submit available non-clinical data while other GLP reproductive toxicology studies were being completed. Based on this submission the FDA removed the clinical hold in March 2021. We are operationally ready and intend to initiate the Phase 2/3 monotherapy MDD trial by the end of March 2021. If positive, the Phase 2/3 trial is intended to serve as one of two registrational trials required by the FDA to support clinical efficacy for the monotherapy treatment of MDD, and we expect topline data in the first half of 2022. In addition to the Phase 2/3 monotherapy trial, we intend to initiate a Phase 2 dose range finding, or DRF, trial for adjunctive treatment of MDD in the third quarter of 2021 to inform dose selection for Phase 3 and provide controlled data to support advancing a Phase 3 adjunctive MDD trial.

Patients in the Phase 2/3 trial will be randomized 1:1 to receive nightly bedtime doses of PRAX-114 or placebo for 28 days in a fully outpatient setting, with two-weeks of additional follow up after the end of the active treatment period. Patients will be required to be between the ages of 18 and 65, have a diagnosis of MDD with a current episode of at least 8 weeks and not more than 24 months in duration, have a HAM-D total score of 23 or higher consistent with moderate to severe MDD, and have had at least one prior episode of MDD . Participants will be excluded if they are currently being treated with an antidepressant, have demonstrated an inadequate response to antidepressant treatment in the current episode or have treatment resistant depression, or if they have comorbid medical or psychiatric conditions that could interfere with the scientific objectives or safety of the trial. The primary efficacy endpoint will be change in the HAM-D total score from baseline at Day 15. A key secondary endpoint will

Table of Contents

be change in the HAM-D score after 28 days of treatment to assess the durability of effect of PRAX-114, and we will also evaluate changes in other depression-related assessments.

As previously described, the initially targeted PRAX-114 oral suspension dose of 45mg was projected to result in the targeted degree of reduction in the HAM-D in patients with MDD. This dose demonstrated rapid and robust improvement in the HAM-D in the Phase 2a MDD trial that was equivalent to the improvement observed in the higher dose groups (60 and 80mg). While all doses were generally well-tolerated and no participants from Part A discontinued from the study due to AEs, the 45 mg dose group demonstrated an optimal benefit/risk profile to guide dose selection for the Phase 2/3 MDD study. These data, in combination with the β-EEG pharmacodynamic biomarker data and the results of the PRAX-114 PK bridging study led to selection of a 40mg PRAX-114 tablet dose for the Phase 2/3 trial. Based on these data, the 40mg PRAX-114 tablet is expected to yield therapeutic exposures that result in effects consistently at or slightly higher than previously seen with the 45mg of the PRAX-114 suspension.

We believe rigor in clinical conduct is essential for a study in MDD to appropriately mitigate historical variability in placebo effect, data reliability and the impact of current social and environmental changes caused by the COVID-19 pandemic. Our operational plan focuses on enrolling the right patients, minimizing placebo response and data variability and ensuring achievement of targeted drug exposure thorough direct verification of study drug adherence. Steps to advance these objectives include:

•Enrollment of patients with moderate to severe MDD and at least one prior episode of MDD (recurrent depression has been associated with a lower placebo response rate).

•A two-level subject and data quality process that includes independent clinical interviews confirming eligibility through the SAFER process and conducting audio confirmation of HAM-D clinical assessments at key timepoints with ongoing rating assessment quality feedback.

•Using sites with a known track-record of high quality data generation and drug-placebo separation in the conduct of MDD trials.

•Integration of a placebo control reminder script at every visit and screening for potential duplicate subjects via a dedicated clinical trial registry.

•Inclusion of the AiCure smartphone-based adherence monitoring system with structured site intervention to address participant adherence issues.

In parallel with the conduct of the Phase 2/3 trial in MDD, an exploratory DRF trial is planned to evaluate additional PRAX-114 doses for inclusion in pivotal Phase 3 studies in MDD. Given that comparable improvement in the HAM-D was observed in all doses (45, 60 & 80mg) examined in the Phase 2a MDD study, this trial will additionally explore the efficacy of lower doses of PRAX-114 to determine the optimal dose range to include in the Phase 3 development program. The planned population for this study is MDD participants who have demonstrated an insufficient response to standard of care antidepressant treatment, based on the following rationale: 1) The preliminary efficacy of PRAX-114 was demonstrated in a MDD population with a majority of participants having demonstrated an insufficient response to standard of care antidepressant treatment, supporting the suitability of this population for efficacy signal detection; 2) to extend these preliminary efficacy findings by generating controlled data for the efficacy of PRAX-114 in the adjunctive MDD population to inform Phase 3 study design in that population; and 3) to support concurrent conduct of the MDD development program studies while realizing data quality and operational efficiency benefits resulting from conducting the DRF study at a subset of the clinical research sites participating in the Phase 2/3 study in MDD.

Patients in this study will be randomized to receive 10, 20, 40, or 60mg adjunctive PRAX-114 or placebo in a 1:1:1:1:1 ratio for 28 days, with 2 weeks of additional follow up after the end of the active treatment period. This trial will enroll approximately 125 patients between the ages of 18 and 65 who are experiencing a current major depressive episode of at least 12 weeks and not more than 24 months in duration, who are being treated with an antidepressant at a stable dose for at least 8 weeks prior to Day 1 and have demonstrated an insufficient clinical response to 1 to 2 adequate trials of antidepressant treatment in the current episode, and who have had at least one prior episode of MDD responsive to antidepressant treatment. Exclusion criteria will be similar to the Phase 2/3 study, with the exception of the insufficient response exclusion. The objective to assess the presence of a dose-response signal for adjunctive PRAX-114 in MDD will use the primary efficacy endpoint of the change from baseline in HAM-D total score at Day 15. Secondary objectives will include evaluating the efficacy for each dose of

Table of Contents

adjunctive PRAX-114 in MDD and the effect of adjunctive PRAX-114 on the HAM-D after 28 days of treatment and impact on other depression-related assessments. This trial will employ the full set of clinical trial quality interventions summarized above for the Phase 2/3 study, and the topline data is planned to be delivered in parallel with the Phase 2/3 monotherapy MDD study.

Upon completion of Part B of our Phase 2a trial, we intend to have a meeting with regulators to discuss further development of PRAX-114 in PMD.

PRAX-114 preclinical data

The goal of our preclinical program was to establish the in-vitro and in-vivo pharmacological profiles, antidepressant potential and tolerability of PRAX-114. In addition, we evaluated translational pharmacodynamic biomarkers to inform clinical development.

Antidepressant activity

To determine the antidepressant-like activity of PRAX-114, we used the Wistar Kyoto, or WKY, rat model. The WKY rat is an inbred rat strain that has increased sensitivity to stress and displays a depressive-like phenotype that is resistant to SSRI and SNRI treatment. A common way to assess depressive-like symptoms in animals is a test known as the forced swim test, or FST. The FST is based on the natural behavior of an animal when placed in a container filled with water from which it cannot escape. The rat will first make efforts to escape by swimming or climbing, but eventually will exhibit floating behavior, which is an indication of behavioral despair and is seen as a surrogate for depression. WKY rats display longer time inactive (floating) over a given time period than normal rats as an indication of increased behavioral despair.

We administered oral doses of PRAX-114 or a placebo to WKY rats and evaluated performance on the FST. At all doses of PRAX-114 tested, 1mg/kg, 3mg/kg and 10mg/kg, we observed a significant reduction in immobility time compared to rats that received a placebo, which we believe reflects an anti-depressive-like reaction or activity of PRAX-114. Importantly, and as described below, at these doses, PRAX-114 did not impair or enhance overall spontaneous activity of the rats in independent assays in the same animals, which we believe indicates that PRAX-114 was generally well-tolerated at these doses.

Tolerability

A common model to assess sedation in rats is the measure of spontaneous locomotor activity, or sLMA. Dosing rats with sedatives dose-dependently reduces their spontaneous movement in this assay. In this model, doses of PRAX-114 up to 30mg/kg had no significant impact on spontaneous locomotion, while doses as low as 1mg/kg had significant antidepressant-like effects in the WKY rat model, demonstrating a wide therapeutic window in these models with preclinical activity at doses well below sedative doses.

We believe that the therapeutic window observed in our in-vivo assays is consistent with the preference of PRAX-114 for extrasynaptic GABAA potentiation observed in-vitro.

EEG as a pharmacodynamic biomarker

In our rat translational biomarker model, we administered PRAX-114 to wild-type rats at doses ranging from 1 to 20mg/kg to assess the impact on power in the beta frequency band. We found that PRAX-114 dose-dependently increased the power in the beta frequency band and these changes correlated with changes in plasma pharmacokinetics. This EEG biomarker was used to inform dose-selection for PRAX-114 clinical studies. In our Phase 1 MAD trial, healthy volunteers administered the 30mg dose of PRAX-114 displayed an approximately 1.6-fold increase in qEEG beta power compared to baseline. In our preclinical studies, doses (and plasma/brain concentrations) that induced a 1.6-fold increase in the beta frequency power in rats were associated with both robust preclinical activity in animal models of depression and anxiety and good tolerability. Specifically, the PRAX-114 dose that is estimated to induce a 1.6-fold increase in EEG beta power activity in rats was efficacious in the rat WKY model of depression and the window between that dose that increased beta power by 1.6-fold increase in EEG and the dose that caused a 50% reduction of spontaneous locomotion in the sLMA sedation assay, or ED50, was ~11-fold, based on brain concentrations. In addition, at this generally well-tolerated brain concentration, PRAX-114 was efficacious in animal models of anxiety including conditional emotional response, or CER, punished drinking, or Vogel, and elevated plus maze, or EPM (Figure 12).

Table of Contents

In the figure below, the lower bound of the preclinical activity in animal models and EEG bars are determined by the brain exposure at the lowest dose at which significant activity was observed (p < 0.05). The lower bound of the tolerability bar represents the TC50 in the brain. The upper bound represents the mean brain concentration at the highest dose tested in a given assay.

Figure 12. Summary of PRAX-114 preclinical data.

Based on our findings in preclinical models, we believe our initial results in humans are supportive of a wide therapeutic window which, in humans, begins at or below a daily dose of 30mg of PRAX-114 and extends to higher doses prior to the onset of potential dose-limiting somnolence or sedation. Our clinical studies to-date suggest that PRAX-114 doses up to 80mg, the highest we have tested in humans, remain generally well-tolerated. We have yet to identify the MTD.

PRAX-944

We are developing PRAX-944, a potentially differentiated selective small molecule inhibitor of T-type calcium channels, for the treatment of ET. We have evaluated the safety and tolerability of PRAX-944 in over 150 healthy volunteers in five separate clinical trials. In these trials, we have studied the safety of PRAX-944 modified release formulation with titration up to 120mg/day and no MTD has been identified. We are currently conducting a Phase 2a proof-of-concept open-label trial in ET patients. Preliminary site data from six participants of the low dose cohort (maximum dose of 40mg daily) showed tremor reduction, which seems to compare favorably to the standard of care agents. We plan to announce topline open-label safety, tolerability and efficacy data, including a high dose cohort, in mid-year 2021.

There is a large body of clinical, preclinical and genetic evidence that points to the involvement of T-type calcium channels in the cerebello-thalamo-cortical, or CTC, circuit, as a main driver of ET. ET is the most common movement disorder, affecting up to seven million patients in the United States, which is seven times more individuals compared to Parkinson’s tremor. ET is a progressive and debilitating movement disorder with action tremors that significantly disrupt daily living. There is a high unmet need for ET patients given the limited treatment options, with only one approved pharmacotherapy that is poorly tolerated, resulting in high discontinuation rates and a small group of patients opting for invasive brain surgeries.

Successful development of T-type calcium channel modulators in ET likely requires a PK profile with a blunted Cmax and thoughtful clinical trial design and endpoint selection. We have designed our development program to include careful selection of clinical endpoints, a modified release formulation and dose titration strategy. We believe the profile of PRAX-944 coupled with its modified release formulation positions it for development as a differentiated therapy in ET.

Table of Contents

Because of the gatekeeper role of T-type calcium channels in regulating neuronal firing patterns in multiple neuronal circuits, we believe PRAX-944 is suitable for potential development across a wide-range of indications in psychiatry and neurology, providing sizable expansion opportunities in addition to ET.

Essential Tremor

ET is the most common movement disorder, characterized by involuntary rhythmic movement in the upper limbs, with or without tremor in other body locations such as the head, vocal cords, or legs. ET is a day-time disease associated with debilitating tremors triggered when a patient voluntarily attempts to move. These tremors significantly disrupt daily living and are progressive in nature, with increases in tremor severity and amplitude commonly observed over the course of the disease. The upper-limb tremor can range from barely visible to greater than 20cm in amplitude.

Unlike Parkinson’s disease, which is characterized by a rest tremor, the tremor of ET occurs with movement and therefore causes direct disability, as people are unable to perform basic, every-day functions such as writing, typing, drinking or feeding themselves. Given the debilitating physical challenges of the disease, ET has also been associated with high prevalence of comorbidities, including anxiety, depression and social phobia.

ET is a clinically well-recognized indication with defined diagnostic criteria established by The International Parkinson and Movement Disorders Society. ET affects between one and two percent of the worldwide population and approximately five percent of adults over 60 years of age. It is estimated that there are up to seven million individuals with ET in the United States, up to seven times more than the second most common movement disorder, Parkinson’s disease.

Despite the prevalence and significant disease burden of ET, only a fraction of patients (an estimated one million based on claims data) are managed with pharmacological therapy, though an estimated 80% of those that are treated discontinue these medications due to limitations in efficacy and tolerability. We believe that the treated population will increase with the availability of new therapies with improved efficacy and tolerability.

Currently, there are only two drugs commonly used in ET. Propranolol, approved by the FDA in 1967, remains the only currently approved therapy for ET in the United States. A non-selective beta blocker, Propranolol is contraindicated for individuals with certain respiratory or cardiac issues, which are common comorbidities in the age group affected by ET. Primidone, an anticonvulsant used off-label, requires slow titration over six to eight weeks and can cause sedation and balance issue while accelerating osteoporosis with long-term use.

As a last line therapy, several thousand ET patients in the United States opt for invasive surgery each year. Interventions include gamma knife and focused ultrasound thalamotomy, where part of the thalamus involved in the CTC circuit is ablated, or deep brain stimulation, or DBS, where an electrode is implanted into the brain. These procedures are generally effective but are associated with significant side effects and risk. Therefore, many patients who are eligible for surgical therapies do not elect to have these procedures.

A significant unmet need remains for the millions of ET patients that are not currently receiving treatment for their ET, or are underserved by existing treatment options. We believe that the relatively concentrated ET treatment setting composed of mainly neurology and movement disorder specialists would allow for the rapid adoption of a new treatment option that offered robust response rates and an improved tolerability profile.

Genetics of Essential Tremor

Our rationale for approaching ET through inhibition of T-type calcium channels is rooted in the genetics of epilepsy. CACNA1G, a gene that encodes for a particular isoform of T-type calcium channels, is one of the most significantly associated genes for generalized genetic epilepsy, or GGE. Some of these epilepsy patients also suffer from comorbid movement disorders such as tremor and ataxia. The odds of observing a T-type calcium channel mutation in the GGE population is 9 times of that of the healthy population. This supports the key role of T-type calcium channels in maintaining excitation and inhibition balance.

Additional human genetic data provide evidence for the role of T-type calcium channels in movement disorders. Whole exome sequencing of early-onset familial ET patients also identified mutations in CACNA1G that segregated with the tremor phenotype in multiple family pedigrees. The importance of T-type Ca++ channels to the function of the CTC circuit is highlighted by variants in the CACNA1G gene which are associated with rare cases of pediatric cerebellar atrophy. Additionally, mutations in the T-type calcium channel have also been reported as

Table of Contents

causative of a form of spinocerebellar ataxia. We believe this genetic link, along with the preclinical and clinical evidence, help confirm the role of T-type calcium channels in the pathophysiology of ET.

Role of T-type calcium channels in ET

T-type calcium channels function as the gatekeepers of neuronal firing patterns, controlling the switch between tonic and burst firing in the CTC circuit. The CTC circuit is a series of brain nuclei or neuron clusters, including the inferior olivary nucleus, cerebellar Purkinje cells, deep cerebellar nuclei, ventral motor thalamus and motor cortex, which work together in regulating coordinated movements and when disrupted generate tremor. All nuclei in this circuit contain pacemaker cells with inherent burst firing capability and express T-type calcium channels, which are known drivers of oscillatory burst firing.

T-type calcium channels are low voltage activated channels that respond to weak depolarization of neuronal membranes and are quickly inactivated (a closed state where the channel cannot be reopened for some time). The opening of T-type calcium channels leads to membrane depolarization, which activates voltage-activated sodium channels, leading to the formation of an action potential and neuronal firing. When only a small number of T-type calcium channels are activated, leading to small T-type calcium channel mediated membrane depolarizations, the neuron generally generates unitary action potentials, also called tonic firing. When the activity of T-type calcium channels is increased, either due to genetic mutations or other changes in network activity that recruit more T-type calcium channels, a longer lasting depolarization is generated, resulting in high-frequency clusters of sodium channel driven action potentials, also called burst firing, as illustrated in the figure below.

Figure 13. T-Type calcium channels are gatekeepers of neuronal firing patterns.

Neuroimaging and neurophysiology studies in ET patients has consistently demonstrated that individual nuclei along the CTC circuit oscillate at the same frequency as the tremor and with strong coherence amongst the brain regions and movement in the affected muscles. Further, intraoperative real-time single-unit recordings of action potentials of individual neurons in the ventral motor thalamus of severe ET patients receiving DBS implants, in periods with and without tremors, further substantiates the central role of the CTC circuit and T-type calcium channels in ET (Figure 14). When no tremor was observed at rest, tonic firing was recorded in neurons of the ventral motor thalamus. During tremor, the same neurons fire in rhythmic bursts that are highly coherent with tremor activity. Furthermore, the emergence of action tremors coincided with the emergence of burst firing. Lesioning or DBS of the ventral motor thalamus has been shown to silence the oscillatory burst firing activity in the CTC circuit, resulting in significant tremor reduction. The strong temporal coordination between the tremors and burst firing, a neuronal firing pattern frequently gated by T-type calcium channel activity, strongly suggest that pharmacological inhibition of these channels may represent an effective pharmacological approach in ET.

Table of Contents

Figure 14. Thalamic neuron burst firing correlated with tremor activity in ET patients.

The role of the CTC circuit and T-type calcium channels has been further confirmed in animal models. A known pharmacological tremor model utilizes administration of harmaline, an alkaloid toxin, to animals. Harmaline, on administration to experimental animals such as rodents, induces an acute action tremor as well as rhythmic burst-firing activity in the CTC circuit similar to that observed in ET patients. We believe this model carries clinically predictive value, as compounds that improve tremor in ET patients clinically (e.g., propranolol, primidone) have also been shown to reduce harmaline-induced tremor preclinically; in contrast compounds that worsen tremor in patients (e.g., caffeine) also worsen tremor in this model. Similar to what’s observed in ET patients, normalizing oscillatory activity in the CTC circuit, for example with DBS, reduces harmaline induced tremor in rodents. Pharmacological inhibition or genetic knockout of T-type calcium channels lead to resistance to harmaline-induced tremor.

PRAX-944 in Essential Tremor

We are advancing a modified release formulation of PRAX-944, a potent and selective small molecule inhibitor of T-type calcium channels, for the treatment of ET.

PRAX-944 Clinical Development in ET

We have evaluated the safety and tolerability of PRAX-944 in over 150 healthy volunteers in five separate clinical trials. In these trials, we have studied the safety of PRAX-944 modified release formulation with titration up to 120mg/day and no MTD has been identified. We are currently conducting a Phase 2a proof-of-concept open-label trial in Australia and New Zealand in ET patients. Preliminary site data from six patients of the low dose cohort showed tremor reduction, which seems to compare favorably to the standard of care agents. We plan to announce topline open-label safety, tolerability and efficacy data, including a high dose cohort, in the first half of 2021.

Phase 1 trials in healthy volunteers using previous IR formulation

An immediate release, or IR, formulation of PRAX-944 was used in the initial Phase 1 trials and reached maximal plasma concentrations within one to three hours of dosing. Adverse events like nausea were associated with the high peak levels. This prompted the development of a modified release, or MR, formulation for PRAX-944 that extends the absorption of the drug over a longer period. We have demonstrated that the MR formulation, which releases approximately 80% of the drug product over seven hours in vitro, reduces the maximum plasma concentration and delays the tmax without significantly impacting the overall AUC. We have also observed that this resulted in improved tolerability relative to an immediate release, or IR, formulation and the ability to sustain targeted therapeutic concentrations. The sustained exposure also enables once daily dosing (Figure 15).

Phase 1 trials in healthy volunteers using MR formulation

In our Phase 1 multiple dose trial of the MR formulation of PRAX-944 in England, doses of 20mg and 40mg were generally well-tolerated over 8 days. Adverse events were transient and occurred at a rate similar to placebo. We observed the following TEAEs all of which were mild to moderate:

•40mg dose (n=6): we observed the nervous system TEAEs of somnolence (33.3%), headache (33.3%) and dizziness (33.3%). The TEAEs also included fatigue (33.3%) and hot flash (33.3%). We observed ECG application site rash, EEG application site skin reaction, nausea, vision blurred, thermal burn (accidental) and euphoric mood in 16.7% of subject each.

Table of Contents

•20mg dose (n=6): we observed the nervous system disorder TEAEs of somnolence (16.7%) and headache (33.3%). We also observed nausea (33.3%), fatigue (16.7%), vomiting (16.7%) and dry throat (16.7%).

•Placebo (n=4): we observed the nervous system TEAEs of headache (25%), somnolence (25%) and dizziness (25%). We also observed fatigue (50%) and nausea (25%).

A single dose of 60mg was not tolerated in the single dose trial due to reports of nausea in five of six subjects and vomiting in three of six subjects. In the multiple dose 20mg and 40mg groups, three subjects reported nausea with one subject also reporting vomiting; these events were mild in severity and resolved on Day 1 of dosing. No subjects reported nausea or vomiting after Day 1 of dosing. While a single dose of 60mg was not well-tolerated, the average peak drug levels (138ng/mL) observed in the 40mg group after eight days of treatment were greater than those seen with the single 60mg dose (130ng/mL) on Day 1. Improved tolerability at higher concentrations following repeated dosing suggests that titration to higher doses might be a viable strategy to further improve the tolerability profile.

Figure 15. Sustained exposures were observed for the MR formulation of PRAX-944.

Quantitative EEG studies in healthy volunteers were used to assess the pharmacodynamic effect of PRAX-944 on T-type calcium channels in the brain. One frequency band known to be driven by T-type calcium channel activation is the sigma frequency band (11 to 15 Hz) during non-rapid eye movement sleep, or NREM sleep. T-type calcium channels expressed in thalamic neurons are critically involved in the generation and modulation of these rhythmic thalamocortical oscillations during NREM sleep.

In our preclinical studies, dosing of normal rats with PRAX-944 led to robust and dose-dependent changes in EEG activity. Because similarly robust sigma frequency band changes after dosing with PRAX-944 are observed during NREM sleep in rats and humans, our hypothesis is that the inhibition of this EEG signal can be used as a pharmacodynamic biomarker. Because the doses at which EEG changes observed in rats are similar to those that demonstrated activity in a preclinical model of essential tremor, or the harmaline model, we believe that this EEG biomarker can be used to estimate the dose of PRAX-944 that will produce a therapeutic effect in ET.

In our Phase 1 trial, 20mg and 40mg doses of PRAX-944 administered to healthy volunteers produced changes in the qEEG recordings of the sigma frequency band during NREM sleep consistent with those observed in rats (Figure 16). This indicated that PRAX-944 reached target levels in the brain needed to inhibit T-type calcium channels. Based on the overlap of these EEG changes with drug levels showing activity in the preclinical harmaline model, we believe that 20mg and 40mg doses of PRAX-944, which were generally well-tolerated in healthy volunteers without titration, have the potential to reduce tremor in patients with ET.

Table of Contents

Figure 16. Exposures of PRAX-944 that decreased sigma band activity were generally well-tolerated in healthy volunteers and were associated with harmaline tremor reduction.

Titration trial in healthy volunteers

Considering that improved tolerability is the key unmet need in ET and that existing data suggest that titration is a viable strategy to further improve PRAX-944’s tolerability profile, we explored titration in a two-part healthy volunteer study. In Part A, a new 5mg PRAX-944 tablet was assessed for low-dose pharmacodynamic effects at in an open-label titration paradigm from 5mg up to 20mg daily. In Part B, PRAX-944 was titrated from 20mg up to 120mg daily, to assess the safety, tolerability and pharmacodynamic activity of higher doses. Participants were randomized to PRAX-944 or placebo in a 3:1 ratio, starting at 20mg daily in the morning and titrated at 20mg increments up to 120mg daily with up to one week in between each dose increment to achieve steady-state plasma concentrations and for the collection of safety data. The total dosing duration was 31 days.

Dosing has completed in this study. The safety data demonstrated that with titration, PRAX-944 was generally well-tolerated up to 120mg daily. There were no SAEs and no severe AEs. The majority of AEs were mild, transient and resolved without intervention (Figure 17). There were no treatment related ECG or EEG abnormalities. Safety laboratory values have generally been within normal limits and there have been no dose dependent excursion from the normal range. Only 1 of 12 participants randomized to PRAX-944 discontinued for a treatment-related adverse event. This participant dropped out after 1 dose (20mg) due to symptoms the participant described as similar to a prior panic attack (not reported at screening). Vital signs, physical examination, clinical laboratory tests, and ECG parameters were all within normal limits. The symptoms self-resolved.

Importantly, no MTD was identified. We believe this provides an advantage for PRAX-944, as it enables a wide dose range for optimizing efficacy and tolerability profile tailored to the key unmet need in ET as well as individual patient needs. Preliminary analyses confirmed the pharmacodynamic changes seen in the previous Phase 1 trial described above, and we will further assess changes across the dose range up to 120mg/day.

Table of Contents

Figure 17. TEAEs Occurring in at least two participants in a dose group or overall.

Phase 2a trial inpatients with ET

We are currently conducting a Phase 2a proof-of-concept open-label trial in up to 12 patients per cohort with ET in Australia and New Zealand. In the first cohort, participants received 20mg daily dosing of PRAX-944 for one week followed by 40mg daily dosing for the second week, taken in the morning. Based on the data from the titration trial in healthy volunteers, we are enrolling an additional cohort of ET patients dosing up to 120mg.

Studies in movement disorders require careful attention to methods for obtaining and scoring outcome measures. We are measuring changes in tremor with different, complementary approaches including components of the Essential Tremor Rating Scale, or TETRAS, Performance Scale and accelerometry. TETRAS is a widely used clinical rating scale that measures the severity of ET. It was based on similar clinical scales which have been used to support regulatory approval of neurosurgical treatments for severe ET. TETRAS has shown good measurement properties and dynamic range compared to other scales.

We are using change from baseline in the rating of upper limb, or UL, items of the TETRAS as the primary efficacy outcome in this proof-of-concept trial because all ET patients suffer from UL tremor. As the UL items drive most of the score on the overall TETRAS and are more reliably rated than other items on the scale, they are therefore expected to have the best signal to noise ratio. UL items have also been the basis of the most recent regulatory approval of neurosurgical treatments for severe ET. We have established rigorous procedures for training and for blinded scoring of efficacy, including using centralized video assessment as an exploratory endpoint, with randomization of the videos and masking to allow for rating concordance. We have also included the overall TETRAS performance scale, or TETRAS-PS, (both site and central video rating) and Kinesia ONE accelerometer, clinical global impression of severity and improvement, or CGI-S and CGI-I, respectively, and the patient global impression of change, or PGI-C, as secondary endpoints in the current open-label study to assess consistency of response across different endpoints.

In this trial, we are enrolling participants with well-established ET, as defined by the Movement Disorders Society, or MDS, Task Force for Tremor as an isolated tremor syndrome of bilateral UL action tremor with at least 3 years’ duration. Patients are required to have a combined bilateral score of ≥10 on the TETRAS UL items as confirmed by site investigator and central video review. This requirement for moderate to severe symptoms at baseline provides a clear and measurable dynamic range for detecting a treatment response. In the first cohort, tremor severity will be evaluated before drug administration, after daily morning dosing of PRAX-944 20mg for 7 days (Day 7), following daily administration of PRAX-944 40mg for seven additional days (Day 14) and one week after administration of PRAX-944 has been stopped (Day 21).

Table of Contents

Preliminary data are available from six participants who completed the trial and received PRAX-944 doses of 20mg followed by 40mg, each for seven days (Figure 18). Preliminary site (primary endpoint measure) and central video assessments of participants TETRAS-PS in this cohort showed generally stable tremor severity between screening and baseline visits. The primary endpoint change suggests dose dependent tremor reduction on the TETRAS-UL which compares favorably to the standard of care agents, and the change was consistent with the central video assessment. Placebo effects in tremor trials are typically low. A recent trial in ET demonstrated a 0.1% effect on the clinical rating scale in the control group. Importantly, five of the six participants remained on propranolol in this study, suggesting that PRAX-944 could also be efficacious as an adjunctive treatment. Similar patterns of improvement were also observed in the full TETRAS-PS, and Kinesia ONE accelerometry scores. The site and central ratings were strongly correlated on the TETRAS-UL and TETRAS-PS with r values of 0.8 and 0.83, respectively.

*As TETRAS PS items are rated on a logarithmic scale, the Weber-Fechner law was used to calculate the percent change in tremor amplitude according to the equation presented in Elble (2018).

Figure 18. Change from baseline in TETRAS scores and percent change from baseline in tremor amplitude as measured by site ratings of the TETRAS PS and Upper Limb subscale in Part A of the ET OL study (N=6).

Table of Contents

A total of 7 participants have completed Part A of this trial. Participants were administered 20mg PRAX-944 QAM for 7 days followed by 40mg QAM for 7 days. These dose levels have been generally well-tolerated. No SAEs and no severe AEs have been observed. The majority of AEs have been mild, transient and resolved without intervention. Six out of 7 participants completed dosing per protocol. One participant discontinued on Day 8 due to anxiety. This participant was also non-compliant with the protocol, stopping propranolol on Day 3 of dosing without consulting study staff.Due to this protocol deviation which would have impacted this participant’s TETRAS scores, this participant is included in safety data but not in efficacy data. No clinically significant ECG or laboratory abnormalities have been reported.

Based on the observed safety profile in the titration healthy volunteer trial and the safety and preliminary activity observed in Part A of the ET trial, we plan to include a high dose cohort (Part B) of up to 12 participants in this on-going ETtrial titrating dosing up to 120mg in an open-label fashion. We also are including a randomized, double-blind, placebo-controlled withdrawal phase to the trial, where participants will either be maintained on their final open-label dose or switched to placebo. The goals of the randomized withdrawal are to obtain blinded confirmation of effect from the open-label titration and to assess for durability of effect.

We plan to announce topline open-label safety, tolerability and efficacy data, including the high dose cohort, from this Phase 2a trial in mid-year 2021.

Additional studies planned for the PRAX-944 program

We plan to initiate a Phase 1 study to explore shorter titration schemes in mid-year 2021. In addition, we plan to initiate a Phase 2b randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of PRAX-944 in the treatment of ET patients in the fourth quarter of 2021.

Preclinical support for advancing PRAX-944

PRAX-944 has been shown preclinically to inhibit all three human T-type calcium channel isoforms, CaV3.1, CaV3.2 and CaV3.3, and has demonstrated high selectivity against L and N-type calcium channels, or CaV1.2 and CaV2.2, respectively, and other key ion channels important for normal physiology, such as the cardiac potassium channel human Ether-à-go-go-Related Gene, or hERG, and the voltage gated sodium channel NaV1.5. Robust selectivity and potency have been demonstrated across both exogenously expressed recombinant channels in a human cell line and naïve channels in isolated dorsal root ganglion, or DRG, neurons from rats using electrophysiological techniques.

HEK CELLS RAT DRG NEURONS

Channel IC50 (nM) Channel IC50 (nM)

Table 19. PRAX-944 is a potent and selective inhibitor of T-type calcium channels.

Consistent with the gatekeeper role of T-type calcium channels in neuronal firing patterns, a gain of function mutation of the T-type calcium channel CaV3.2 leads to pathological burst firing in thalamic neurons in a rat model known as the GAERS model. Administration of PRAX-944 resulted in complete suppression of the pathological burst-firing in thalamic neurons derived from the GAERS model.

We validated the therapeutic potential of PRAX-944 to treat ET using the harmaline-induced tremor model in rats. Administration of harmaline triggers ET-like tremors in experimental animals as well as pathological burst firing throughout the CTC circuit. We observed a large and dose-dependent decrease of harmaline-induced tremor in rats treated with PRAX-944 as compared to vehicle-treated animals, when measured as % increase of tremor from pre-

Table of Contents

harmaline baseline. This result served to both demonstrate the potential of PRAX-944 in ET and as independent evidence of the critical role of T-type calcium channels in tremor reduction.

Figure 20. PRAX-944 led to a dose-dependent inhibition of tremors in the rat harmaline model.

EEG as a pharmacodynamic biomarker for dose selection

PRAX-944 robustly and dose-dependently decreased EEG power in the sigma frequency band during NREM sleep in rats. The effect of PRAX-944 on the EEG observed in rats when dosed with PRAX-944 indicates its ability to mediate the blockade of T-type calcium channels in the thalamocortical circuit, suggesting that this effect is a pharmacodynamic biomarker for PRAX-944. Because the doses at which the EEG changes are observed are similar to those that demonstrate tremor reduction in the harmaline model, we believe that this biomarker can be used to estimate the dose that could be effective in treating ET.

Figure 21. PRAX-944 decreased EEG sigma power during NREM sleep in rats.

RARE DISEASE PORTFOLIO

We are advancing several programs which we believe offer significant therapeutic benefits for rare disease populations over the current standard of care. We believe that all of the programs in our rare disease portfolio have the potential to be differentiated molecules, have foundational underpinnings in human genetics, utilize translational biomarker tools and have the potential for early clinical signal detection.

Table of Contents

Our most advanced rare disease candidate is PRAX-562, which is currently in a Phase 1 trial in adult healthy volunteers in Australia. Its mechanism suggests that PRAX-562 has broad potential to treat many diseases of neuronal hyperexcitability. We are pursuing development in a subset of rare devastating diseases, initially rare adult cephalgias and pediatric epilepsies.

The remainder of our current rare disease portfolio consists of precision medicines approaches to address genetically defined populations suffering from Developmental and Epileptic Encephalopathies, or DEEs. DEEs are rare neurologic diseases characterized by early onset (< 2 years of age), frequent seizures, abnormal epileptiform electroencephalographic activity, developmental impairment and resistance to available antiepileptic drugs. Furthermore, DEEs are associated with a high mortality rate and comorbidities such as developmental delay in addition to behavioral disorders, movement disorders, pain and sensory dysfunction and sleep disruptions.

The understanding of the etiology of DEEs has been revolutionized by recent whole-exome sequencing initiatives that showed over 60 genetic causes of epilepsy. An underlying pathologic feature of many DEEs is the dysregulated neuronal activity leading to hyperexcitability, seizures and associated comorbidities. This phenomenon is observed in many pediatric DEEs with an identified genetic cause, such as SCN8A, SCN2A, KCNT1, KCNQ2, KCNQ1, STXBP1 and SYNGAP1 epilepsy, as well as epilepsies in which a genetic cause remains unclear, such as Lennox Gastaut Syndrome, or Doose Syndrome. Up to 40% of DEEs are caused by single gene mutations, enabling precision medicine approaches.

Our lead precision medicine candidate is PRAX-222, an antisense oligonucleotide, or ASO, for lowering the expression levels of the protein encoded by the gene SCN2A, in patients with gain-of-function, or GOF, mutations in SCN2A, the underlying cause of SCN2A GOF DEE. We have also entered into a research collaboration with The Florey Institute of Neuroscience and Mental Health to develop three novel ASO therapies for the treatment of patients with SCN2A loss-of-function mutations and two additional rare epilepsy targets. This partnership positions us at the forefront in rare epilepsy drug development with six distinct programs for the treatment of six different rare epilepsies.

Given the overlapping biology, phenotypic presentation and clinical execution considerations, we believe that developing a portfolio of drugs to treat DEEs will create a distinct body of knowledge and operational synergies across our rare disease portfolio, positioning us as a leader in developing meaningful therapies for this group of patients with devastating unmet clinical needs.

PRAX-562

Standard of care sodium channel blockers, such as Tegretol (carbamazepine), Lamictal (lamotrigine), Dilantin (phenytoin) and many others are an important class of medicines in neurology and psychiatry. All standard of care sodium channel blockers modulate neuronal activity by targeting peak sodium current, which can reverse the pathological neuronal hyperexcitability that underlies many CNS conditions, but simultaneously affects the physiological cellular action potential firing required for a functioning nervous and cardiovascular system. Hence, this class is widely used for the treatment of epilepsy, pain, migraine and bipolar disorder. However, the efficacy of sodium channel blockers is generally limited by side effects, many attributable to on-target toxicological effects.

PRAX-562 is designed as the first selective, persistent sodium current blocker that has the potential of reducing pathological neuronal hyperexcitability with an improved tolerability profile. PRAX-562 is in development for the treatment of a broad range of rare, devastating CNS disorders, such as severe pediatric epilepsies and adult cephalgia. We intend to pursue development in a subset of rare cephalgias, initially Short-lasting Unilateral Neuralgiform headache attacks with Conjunctival injection and Tearing, or SUNCT, Short-lasting Unilateral Neuralgiform headache with Autonomic symptoms, or SUNA, and Trigeminal Neuralgia, or TN, along with pediatric epilepsies.

In in-vitro studies, PRAX-562 selectively blocks persistent sodium current across all subtypes of sodium channels with minimal effects on the peak sodium current that is critical for the normal physiological function of these channels. In line with this selectivity for persistent current, PRAX-562 has been shown in ex-vivo studies to reduce neuronal hyperexcitability without impairing normal neuronal function. This is in contrast to marketed sodium channel blockers which significantly impact normal neuronal function, leading to a narrow therapeutic index.

To date, PRAX-562 has demonstrated robust pharmacological activity in a preclinical in-vivo seizure model with significantly improved tolerability compared to other sodium channel blockers, suggesting a potentially improved therapeutic index. The characteristics of PRAX-562 are expected to make it a versatile molecule that we

Table of Contents

believe can be broadly applied in diseases of hyperexcitability where sodium channel blockers have demonstrated efficacy but poor tolerability. Recent preclinical studies with PRAX-562 have demonstrated a dose-dependent and complete prevention of spontaneous seizures in a mouse model of SCN2A GOF DEE and evoked seizures in a human knock-in mouse model of SCN8A DEE.

We have initiated a Phase 1 trial of PRAX-562 in Australia to evaluate the safety, tolerability, PK and effects on an exploratory EEG biomarker in up to 129 adult healthy volunteers. This trial has successfully completed the SAD portion up to the maximum planned dose, with no dose limiting toxicities, and has advanced to MAD cohorts. Maximum observed PRAX-562 plasma concentrations observed to date, at Cmax in the SAD study and at trough levels in the MAD study, exceed those at the effective concentration 50, or EC50, in the MES mouse model, a model with good predictive validity, without dose limiting toxicities. We are currently at the highest preplanned dose of PRAX-562 in our MAD trial and intend to escalate further if it continues to be generally well-tolerated. We anticipate the initial proof-of-concept trial in rare adult cephalgia patients to initiate in the second half of 2021.

The clinical development plan for PRAX-562 encompasses exploring the broad potential for the mechanism of action in rare diseases through proof-of-concept trials in rare adult cephalgia patients and then expanding into a range of rare pediatric DEEs. The FDA has granted rare pediatric disease designation for PRAX-562 for the treatment of SCN2A and SCN8A developmental and epileptic encephalopathies, or SCN2A-DEE and SCN8A-DEE, respectively.

Voltage-gated sodium channels, persistent sodium current and neuronal excitability

Voltage-gated sodium channels, or VGSCs, are transmembrane proteins that are required for electrical signaling and therefore communication in neurons. VGSCs respond to changes in the membrane potential and are tightly regulated by their biophysical properties. Upon opening of VGSCs, sodium ions can move into the cell leading to a depolarization and therefore excitation of the neuron. This sodium current is the initiator and driver of neuronal action potentials, or APs, the primary means of electrical signal propagation along the neuron’s axon.

The family of VGSCs consists of nine highly related isoforms (NaV1.1 – NaV1.9) with differential tissue distributions and functions. NaV1.1, 1.2 and 1.6 are the major sodium channels expressed in the central nervous system.

Isoform Gene Expression

NaV1.1 SCN1A CNS

NaV1.2 SCN2A CNS

NaV1.3 SCN3A CNS/Pancreas

NaV1.4 SCN4A Muscle

NaV1.5 SCN5A Heart

NaV1.6 SCN8A CNS/PNS

NaV1.7 SCN9A PNS

_______________________________

CNS: Central Nervous System, PNS: Peripheral Nervous System

Table 22. Sodium Channel Isoforms and tissue distribution.

VGSCs undergo a structural change that alter their ability to conduct sodium ions (Table 22) and are triggered to open upon excitation, or depolarization, of the cell membrane allowing sodium ions to enter the neuron. Sodium influx further excites, or depolarizes, the neuron, leading to the opening of even more sodium channels. This series of events can lead to a large peak sodium current underlying the initiation and propagation of neuronal action potentials, or APs, the primary means by which neurons propagate information in the nervous system. To prevent overexcitation of neurons, or hyperexcitability in the form of excessive high frequency AP firing, the majority of sodium channels only open very briefly after activation (1-2ms), followed by a refractory period of inactivation or non-responsiveness.

Table of Contents

However, at membrane potentials below the AP firing threshold, a small subset of sodium channels can remain open for hundreds of milliseconds, carrying the so-called persistent sodium current. Persistent sodium current is present under physiologic conditions where it modulates excitability of neurons and can be significantly increased in pathologic states (Figure 23).

Figure 23. Impact of a pathological mutation on sodium channels.

There are currently more than 15 sodium channel blockers in the market commonly used to treat diseases such as epilepsy, bipolar disorder and pain. While standard of care sodium channel blockers, such as carbamazepine, lamotrigine and phenytoin, inhibit persistent sodium current, they likely also block peak sodium current at therapeutic concentrations, which can cause significant adverse events such as ataxia, drowsiness and dizziness, and therefore have a very narrow Therapeutic Index, or TI.

Genetics of persistent sodium current

In published whole-exome sequencing studies of diverse patient populations, mutations in all voltage gated sodium channel subtypes have been reported as a likely cause of disease. Furthermore, gain-of-function mutations that are associated with disease can cause an increase of persistent sodium current, raising the idea that this might be a critical driver of hyperexcitability in neurologic disorders.

The specific disease that a patient develops depends on both the sodium channel subtype and where the affected sodium channel is expressed. Gain-of-function mutations in SCN2A, or NaV1.2, and SCN8A, or NaV1.6, two of the major sodium channels in the brain, cause early onset epileptic encephalopathies with frequent seizures and developmental delay. Gain-of-function mutations in NaV1.1, NaV1.4, NaV1.5 and NaV1.7 cause familial hemiplegic migraine, myotonia, cardiac arrythmia and severe pain disorders, based on their primary expression in the CNS, muscle, heart and pain pathways, respectively. These channelopathies demonstrate the important role persistent sodium current plays as a modulator of cellular excitability.

Our initial indications for PRAX-562

Developmental and Epileptic Encephalopathy

Approximately 100,000 children suffer from DEEs in the United States alone, with over two hundred thousand children affected world-wide. An underlying pathologic feature of many DEEs with both known and unknown genetic causes, is the dysregulated neuronal activity leading to hyperexcitability and subsequently to seizure.

Sodium channel blockers have been a critical component of the pharmacological management of seizure related conditions, including epilepsy, for decades. However, current standard of care sodium channel blockers are limited by a narrow therapeutic window and inadequate efficacy. We believe these limitations are largely due to blockage of peak sodium current and disruption of normal neuronal function at or near therapeutic doses and significant off-target activity.

Table of Contents

Given the role of persistent current in modulating excitability, we believe that PRAX-562 has the potential to be a broadly efficacious and generally well-tolerated antiepileptic drug for the treatment of DEEs of both genetic and unknown etiology.

Cephalgia

SUNCT, SUNA and TN are devastating headache disorders with limited treatment options. SUNCT and SUNA are part of a specific class of cephalgias known as Short Lasting Unilateral Neuralgiform headaches. These headaches are characterized by severe burning, stabbing and electrical unilateral head pain that is typically 9 to 10 in the Visual Analogue Scale, or VAS, for pain. These headache attacks last between one second and ten minutes in duration and can occur up to 600 times per day. SUNCT and SUNA headaches are rare diseases with a prevalence estimated to be 6.6 per 100,000 based on a recent Australian study.

SUNCT and SUNA are often refractory to standard migraine and headache treatments, but are highly responsive to intravenous, or IV, infusion of the sodium channel blocker lidocaine. Response under IV lidocaine requires continuous infusion in an inpatient setting and is associated with side effects such as nausea, vomiting and cardiovascular effects, with headache attacks returning in majority of patients within days of IV lidocaine withdrawal. Preventative treatment of SUNCT and SUNA often includes oral sodium channel blocker lamotrigine, but this is limited by partial efficacy, tolerability concerns and the requirement of several weeks of dose-titration to reach therapeutic doses.

TN, also known as tic douloureux, is characterized by intense, stabbing, electric-shock pain typically in the lower face and jaw, usually on one side of the face. TN is thought to be caused by irritation of the trigeminal nerve.The pain can be triggered by an action as simple as washing or touching the face smiling or talking. These attacks can progressively worsen over time, especially if left untreated. Epidemiologic data are variable across the world and depend on definitions used for diagnosis, with reported annual incidence ranges from 4.3 to 26.8 per 100,000.

Anticonvulsive medications that modulate voltage-gated sodium channels, such as carbamazepine, are considered first line treatment for TN. Additional therapies used to treat TN include other medications (e.g., gabapentin, baclofen, amitriptyline, topiramate), as well as various procedures or surgical interventions (such as vascular decompression or gamma knife). These therapies are reported to all have inconsistent results in controlling the pain, with only about 50% success over time.

The limited FDA-approved treatments specific to SUNA or TN, or complete absence of FDA-approved treatments specific to SUNCT, combined with high comorbidity and healthcare utilization, substantiates the need for an efficacious, generally well-tolerated and orally bioavailable sodium channel blocker to treat headache attacks in acute and preventative settings.

PRAX-562 preclinical data

PRAX-562 is a highly differentiated, potent and selective inhibitor of persistent sodium current designed to overcome the limitations of currently available sodium channel blockers. PRAX-562 preclinical studies were designed to test our belief that the block of persistent sodium current is sufficient to demonstrate robust activity in animal models of hyperexcitation and that the selective block of persistent sodium current over physiological peak current leads to an improved therapeutic index.

Selective inhibition of persistent sodium channels

In preclinical studies, PRAX-562 is a highly potent inhibitor of persistent sodium current as measured in cell-based assays, in which sodium channel isoforms are heterologously expressed and channel activity is measured via patch clamp electrophysiology. Using electrophysiological voltage protocols, the effect of compounds on a specific channel state (e.g., peak current vs persistent current) can be measured. When compared to other approved sodium channel inhibitors for various neurological indications, PRAX-562 was hundreds of times more potent at inhibiting persistent sodium current. PRAX-562 had an IC50 of 141 nM compared to SOC sodium channel blockers lamotrigine and carbamazepine which had an IC50 of 78,530 nM and 77,520 nM, respectively – a potency difference of over 500-fold. PRAX-562 was ~60 fold selective for inhibiting persistent current over peak current.

Table of Contents

Figure 24. PRAX-562 is approximately 150-fold more potent and two to nine-fold more selective for persistent sodium current than standard sodium channel blockers.

The selective block of persistent sodium current reduces neuronal hyperexcitability without affecting the action potential, or AP, amplitude, which is required for normal neuron function. In mouse brain slice experiments, a hyperexcitable state can be mimicked by artificially depolarizing the neuron using the patch clamp method, which elicits high frequency AP firing. PRAX-562 reduced the neuronal AP firing frequency, an indicator of neuronal excitability, without a significant effect on AP amplitude, an indicator of normal neuronal function, suggesting reduction of hyperexcitability without impacting the ability of the neuron to respond to physiologic stimuli. In comparison, carbamazepine, a SOC sodium channel blocker, at comparable concentrations (relative to the potency in cells heterologously expressing NaV1.6), excessively decreased AP firing almost completely and reduced the amplitude of APs, indicating impairment of normal function.

Table of Contents

PRAX-562 Representative AP Traces Carbamazepine Representative AP Traces

Figure 25. PRAX-562 reduced neuronal hyperexcitability (AP frequency) without impairing normal function (minimal effect on AP amplitude). In contrast, carbamazepine significantly reduced the AP amplitude suggesting impairment of normal function.

Preclinical in vivo pharmacological activity, tolerability and EEG pharmacodynamic biomarker

We investigated the preclinical activity of PRAX-562 in a maximal electroshock model of epilepsy, or MES model, that has shown good predictive validity for clinical anti-convulsant activity, and compared it to the effects of SOC sodium channel blockers carbamazepine and lamotrigine. To determine how well PRAX-562 is tolerated, we compared its effects on spontaneous locomotor activity, or sLMA, to the effects of carbamazepine and lamotrigine.

PRAX-562 was able to block seizures completely in mice at a dose that does not impair locomotor function (10mg/kg). In contrast, carbamazepine and lamotrigine only achieve full block of seizures in this model at doses that also show impairment of locomotion. PRAX-562 at a dose of 2mg/kg, inhibited the epilepsy response to half of its maximum value, or ED50. Inhibition of sLMA required an estimated dose of 44mg/kg to obtain 50 percent inhibition, or TD50.

Table of Contents

Figure 26. Doses of PRAX-562 resulting in potent anticonvulsant activity were associated with minimal effects on general locomotor activity.

We calculated the therapeutic index, or TI, of each molecule as the preclinical tolerability/pharmacological activity ratio. This ratio is calculated by dividing the plasma and brain concentrations at the dose that reduces locomotion by 50% by the concentrations that reduce seizures by 50%. We found that PRAX-562 had a significantly improved therapeutic index of ~16 fold (based on brain concentrations) and ~17 fold (based on plasma concentrations) compared to the currently prescribed sodium channel blockers carbamazepine and lamotrigine, which had a much lower protective index of three to six-fold. Notably, standard sodium channel blockers, such as carbamazepine and phenytoin, show severe toxicity in humans at exposures that are only about 1.5 to 3 times the target therapeutic exposures, underscoring the need for modulators of sodium channels with an improved tolerability.

Molecule Plasma TherapeuticIndex Brain TherapeuticIndex

Carbamazepine 3.4 x 5.9 x

Lamotrigine 6.4 x 4.6 x

Therapeutic Index (TI) = TC50/ EC50

Table 27. Compared to lamotrigine and carbamazepine, PRAX-562 had an increased ratio between drug levels that demonstrated preclinical pharmacological activity

versus those that caused toxicity.

The auditory steady state response, or ASSR, is a non-invasive EEG measure of excitatory/inhibitory balance in the brain. This response is elicited with short lasting (2sec) auditory stimuli that lead to brain activity changes that are measured as a 40Hz EEG signature and depend on network activity between excitatory and inhibitory cortical neurons. We believe that persistent current block has the potential to lead to reduced excitability of the network and will be measurable with this endpoint.

Table of Contents

Consistent with this hypothesis, dosing normal mice with PRAX-562 led to a dose-dependent decrease in the ASSR amplitude (40Hz power). This effect was maximal at doses that have robust anticonvulsant effects in the maximal electroshock model.

Figure 28. PRAX-562 dose-dependently reduced the 40Hz EEG power of the auditory steady state response in mice.

Together, our data suggest that the selective effects of PRAX-562 on hyperexcitable states without affecting normal neuronal function led to the robust preclinical reduction of seizures and improved tolerability seen in animal models. As shown below, exposures of PRAX-562 that led to biomarker change (ASSR amplitude reduction shown in top row) also demonstrated robust anticonvulsant activity (shown in middle row). Moreover, PRAX-562 has a ~16.4 fold protective index based on the spontaneous locomotor activity (shown in bottom row), which is a significant improvement over reported effects of approved sodium channel blockers. In the figure below, the lower bound of the preclinical pharmacological activity range, EEG and tolerability bars is determined by the brain EC50 (preclinical seizure and ASSR assays) or TC50 (tolerability assay) in a given assay and the upper bound represents the mean brain concentration at the highest dose tested in a given assay.

Figure 29. Summary of PRAX-562 preclinical data.

Recent preclinical studies with PRAX-562 have demonstrated a dose-dependent and complete prevention of spontaneous seizures in a mouse model of SCN2A GOF DEE and evoked seizures in a human knock-in mouse model of SCN8A DEE.

Table of Contents

We believe that the profile of PRAX-562 may translate into therapies with the potential for clinical efficacy and tolerability across several indications caused by underlying hyperexcitability where standard sodium channel blockers have shown efficacy, albeit with limited tolerability, such as rare pediatric epilepsies and cephalgia like SUNCT/SUNA and TN.

PRAX-562 clinical development in cephalgia and DEEs

We have initiated a randomized, double-blinded Phase 1 trial in Australia to evaluate the safety, tolerability and pharmacokinetics of single and multiple ascending doses of PRAX-562 in up to 129 adult healthy volunteers between the ages of 18 and 55. In addition, we are using ASSR as an exploratory pharmacodynamic biomarker in this trial to determine the doses required to achieve pharmacological blockade of persistent sodium current, which we believe is a potential indicator of efficacy in patients. Preliminary analysis of the completed SAD cohorts indicate that PRAX-562 is generally well-tolerated up to the maximum planned dose. Safety data reviewed included adverse events, vital signs, ECG, C-SSRS, physical examination and safety laboratory data. There have been no reported SAEs, severe AEs or any AEs leading to study withdrawal or discontinuation as of the safety review committee meeting convened on February 18, 2021. The current trial has advanced to MAD evaluations and maximum observed plasma concentrations achieved in healthy normal subjects, at trough levels, exceed those at the effective concentration 50 (EC50), in the MES mouse model. We are currently at the highest preplanned dose of PRAX-562 in our MAD trial and intend to escalate further if it continues to be generally well-tolerated. We anticipate the initial proof-of-concept trial in SUNCT/SUNA and TN patients to initiate in the second half of 2021.

The clinical development plan for PRAX-562 encompasses exploring the broad potential for the mechanism of action in rare diseases through proof-of-concept trials in rare adult cephalgia, then expanding into a range of rare pediatric DEEs. The FDA has granted rare pediatric disease designation for PRAX-562 for the treatment of SCN2A-DEE and SCN8A-DEE.

Along with our PRAX-562 program, we are developing a portfolio of sodium channel blockers that we plan to advance in other rare CNS disorders.

PRAX-222

PRAX US Equity

PRAX · 10-K · period ended 2020-12-31

Item 1. Business

PRAX US Equity☆

PRAX · 10-K · period ended 2020-12-31

Item 1. Business

PRAX US Equity