Function Oncology investment analysis

October 31, 2023

This is not investment advice. We used AI and automated software tools for most of this research. A human formatted the charts based on data / analysis from the software, prompted the AI to do some editing, and did some light manual editing. We did some fact checking but cannot guarantee the accuracy of everything in the article. We do not have a position in or a relationship with the company.

Overview

Function Oncology is a San Diego-based precision medicine firm. They are developing a CRISPR-powered platform to identify personalized cancer treatments. Their approach emphasizes understanding gene function rather than only relying on traditional sequencing to profile cancer.

This approach seeks to uncover therapeutic opportunities by identifying unique drug target dependencies in patient samples, addressing limitations found in conventional next-generation sequencing methods. For example, sequencing-based approaches primarily identify mutated genes with roles in cancer, while Function's approach could also identify non-mutated genes. The company test patient samples for susceptibility to every FDA approved cancer drug. They do this by designing a sgRNA library to inhibit the targets of approved cancer drugs.

Function's platform thus has potential value in identifying repurposing opportunities or expanding the use of existing cancer drugs, identifying potentially beneficial, personalized therapeutics that would otherwise have been overlooked. The platform also could be used for diagnostics, patient stratification for clinical trials, and identifying novel drug targets and developing new drugs.

In April 2023, the company secured $28 million in Series A financing led by a16z and Section 32, with participation from Casdin Capital and Alexandria Venture Investments.

Highlights and risks

Valuation

Given the early stage of the company and limited information about its programs, we did not conduct a valuation analysis.

Platform overview

Function's platform utilizes a CRISPR-powered personalized functional genomics platform, aiming to provide a more patient-specific and detailed profile of cancer.

By using CRISPR/Cas9 to screen for gene dependencies on drug targets, this platform directly assesses how silencing certain genes impacts tumor cell survival or proliferation. Thus, if a tumor cell shows a dependency on a gene targeted by a particular drug, it suggests the tumor may be sensitive to that drug.

The Fx Heme library targets genes of all FDA-approved oncology drugs. This provides a broad overview, potentially highlighting responses to drugs not traditionally associated with a particular cancer type. This approach could also be used for other types of cancer.

Key features include:

• CRISPR-powered Functional Genomics: Unlike traditional gene sequencing methods, this platform focuses on gene function to identify optimal therapeutic opportunities.
• Patient-Specific Insights: By modulating genes using CRISPR directly in patient clinical samples, they can uncover drug target dependencies that might be overlooked by conventional genomic profiling.
• Potential Applications: Insights from this platform can guide tailored treatment strategies and influence the development of novel cancer therapies, including potentially discovering non-mutated "hidden" drug targets

Case study in AML

While limited detail on the platform is available, we can deduce some information from their abstract at AACR 2023. This abstract describes use of the platform to predict clinical response to sorafenib. Given that current AML treatments primarily depend on genomic profiling for molecular lesions, such as the FLT3 mutation, which only partially stratifies the target patient group, there's a clear need for a more personalized approach.

To fill this gap, Function Oncology's platform uses CRISPR/Cas9 technology to systematically characterize gene dependency on drug targets. In their method:

• They collect retrospective pre-therapy samples from AML patients.
• Primary tumor cells from these samples are then transduced with a lentivirus carrying the Cas9 enzyme and a specially-designed sgRNA library called Fx Heme.
• This library is crafted to inhibit the expression of genes related to all FDA-approved oncology drugs.
• After transduction, the cells are harvested at various intervals, and the distribution of sgRNA is determined by amplicon sequencing of DNA barcodes.
• This information allows for the identification of gene dependencies, essentially revealing how specific genes affect drug responses.

Amplicon sequencing refers to the targeted sequencing of specific regions of the genome, in this case, the DNA barcodes associated with each sgRNA. The process involves designing PCR primers to amplify (i.e., create many copies of) these specific DNA regions so that they can be sequenced in depth. By doing so, one can identify and quantify specific genetic elements of interest in a sample.

As cells grow, proliferate, or die in culture, the relative abundance of different sgRNAs in the population will change. If an sgRNA targets a gene that's essential for the AML cells' survival, then cells transduced with that sgRNA will likely die off. As a result, the abundance of that sgRNA's barcode will decrease in subsequent analyses. Conversely, if an sgRNA targets a gene that, when knocked out, gives the AML cells a growth advantage or makes them resistant to a drug, then cells with that sgRNA might proliferate more than others. This would lead to an increase in the abundance of that particular sgRNA's barcode.

The platform's efficacy was evaluated using samples from AML patients who underwent treatment combining the tyrosine kinase inhibitor sorafenib with chemotherapy. From the study of 22 AML patients, the Fx Heme platform showed promising results in predicting clinical outcomes, with 78.6% sensitivity, 87.5% specificity, and 91.7% PPV (Positive Predictive Value). This performance surpassed the stratification based solely on the FLT3 mutation.

Platform in-depth

Below is more detail on the platform, using their AML program as an example:

• Cell Types and Samples:
• Primary tumor cells from AML patients: These cells are crucial as they will be the subjects of the CRISPR-based functional genomic profiling.
• Complexity / cost: Primary cells can be hard to maintain in culture, and their availability may be limited. They also often require specific media and growth factors, which can be costly.
• Scalability: Sourcing primary samples from a large number of patients may be challenging due to ethical considerations and donor variability.
• Technical risk: Ensuring cell viability and preventing contamination are concerns. Additionally, inter-patient variability may introduce inconsistency in results.
• Reagents and Constructs
• Lentivirus harboring Cas9 enzyme: A lentiviral delivery system is used because of its ability to integrate into the host genome and achieve stable expression. The Cas9 enzyme is the protein responsible for generating DNA double-strand breaks at targeted sites.
• sgRNA library: This library contains a collection of specially designed short guide RNAs (sgRNAs) that target specific genes for inhibition. In this study, the library targets genes encoding the proteins of all FDA-approved oncology drugs. The library also includes internal positive and negative control sgRNAs.
• Complexity / cost: Lentiviral production requires specific conditions and can be expensive. Creating a custom sgRNA library, especially one as comprehensive as Fx Heme, is complex and resource-intensive.
• Scalability: Producing large amounts of lentivirus or sgRNAs consistently can be challenging.
• Technical risk: There's a risk of off-target effects with CRISPR. Inefficient lentiviral transduction or sgRNA design flaws can affect the experiment's validity.
• Experiments
• Transduction of AML cells with lentivirus: This step is crucial to introduce the Cas9 enzyme and sgRNA library into the AML cells.
• Harvesting and preparation of cells: After transduction, the cells need to be collected at multiple timepoints to analyze the effects of gene knockouts on cell viability or other phenotypic changes.
• Amplicon sequencing of DNA barcodes: This step identifies which sgRNAs are present in the cell population over time. Changes in sgRNA abundance can provide insights into gene dependencies.
• Complexity / cost: Transduction efficiency can vary, and optimizing conditions for each primary cell sample can be resource-intensive.
• Scalability: Handling multiple patient samples simultaneously requires meticulous tracking to prevent cross-contamination or mix-ups.
• Technical risk: Inconsistent transduction efficiency, variations in cell harvesting times, or issues in amplicon preparation can introduce variability.
• Equipment
• Cell culture equipment: Including incubators, biosafety cabinets, and centrifuges for maintaining and processing the AML cell samples.
• Sequencing machine: For amplicon sequencing of DNA barcodes. Common platforms include Illumina's MiSeq or HiSeq.
• Quantitative PCR machines: Useful for quantifying specific sequences if needed.
• Microscopy: To visualize and assess cell health or phenotype changes if necessary.
• Complexity / cost: Sequencing machines, especially high-throughput ones, are expensive to purchase and maintain.
• Scalability: While sequencing machines can handle multiple samples, increasing throughput requires parallel processing and more machinery.
• Technical risk: Sequencing errors or machine malfunctions can lead to data loss or require repetition of experiments.
• Bioinformatics Tools and Analysis
• Sequence analysis software: To analyze the sequencing data, identify and quantify the sgRNAs, and assess changes in barcode abundance.
• Statistical software: To calculate phenotype scores, sensitivities, specificities, and positive predictive values. This software can also be used to compare dependencies with clinical outcomes.
• Complexity / cost: High-quality bioinformatics tools and expertise are necessary for data analysis. Developing custom tools or pipelines is resource-intensive.
• Scalability: Analyzing large datasets requires significant computational power and storage.
• Technical risk: Errors in data analysis or interpretation can arise from software bugs, flawed algorithms, or human oversight.
• Other Reagents
• Cell culture media, supplements, and reagents: Essential for maintaining the AML cells in culture.
• Lentiviral packaging plasmids: Necessary for producing the lentivirus.
• Reagents for molecular cloning: If there's a need to modify or customize the sgRNA library or other constructs.
• Complexity / cost: Specialized reagents, particularly those needed for primary cell culture or lentiviral production, can be expensive.
• Scalability: Securing a consistent supply of high-quality reagents for large-scale experiments can be challenging.
• Technical risk: Contamination or degradation of reagents can compromise experiment validity.
• Challenges
• Efficiency of transduction: Not all cells may be transduced effectively, which could impact results.
• Off-target effects: CRISPR/Cas9, while precise, can sometimes target unintended regions of the genome.
• Cell viability: The process of transduction and subsequent knockouts might affect the health and viability of the AML cells.
• Misattribution of Therapeutic Effect: Polypharmacology or drugs with poorly characterized targets or mechanism of action creates a risk of misattributing its efficacy. This can lead to incorrect conclusions about the importance of the intended target.
• Confounding Results: If a drug has multiple targets, knocking out a gene encoding one of its targets might not fully recapitulate the drug's effect, leading to potential misinterpretations.
• Need for Further Validation: Findings from such platforms would need further validation, preferably using orthogonal approaches. For example, using small molecules that are highly selective for the identified target or using genetic approaches like RNAi for validation.

The most complex and costly steps are likely the creation and maintenance of the custom sgRNA library, the lentiviral transduction process, and the amplicon sequencing. The bioinformatics analysis is also a critical component, and errors here can invalidate all prior work. Scaling the platform may face challenges, particularly when handling a large number of primary patient samples or increasing the breadth of the sgRNA library.

Potential applications

At its core, the technology provides a method for identifying which drugs could target an individual's tumor. Broadening beyond AML, this can pave the way for truly personalized cancer treatments across a wide variety of cancer types, moving from one-size-fits-all or even mutation-specific treatments to individualized therapy plans based on a tumor's specific gene dependencies. In addition to identifying new therapeutic targets, the platform can also be useful for drug repurposing and developing combination therapies. There are also applications in companion diagnostics, treatment monitoring and adaptive therapies, and clinical trial design.

Predicting response to drugs

Function's platform has the potential to predict which drugs a patient's tumor is likely to respond to. The platform is designed to assess how various drugs would impact tumor cell survival or proliferation, by silencing the genes that express the drug's target (which essentially imitates the effect of inhibiting a target). Thus, if a tumor cell shows a dependency on a gene targeted by a particular drug, it suggests the tumor may be sensitive to that drug.

The Fx Heme library (and presumably other libraries developed by the company) targets genes of all FDA-approved oncology drugs. This provides a broad overview, potentially highlighting responses to drugs not traditionally associated with a particular cancer type.

Several other current and past companies have products for predicting which drugs a tumor will respond.

Current solutions for predicting tumor response include:

• Genomic Profiling (e.g., FoundationOne CDx, Tempus xT)
• Application: These tests identify mutations, amplifications, or fusions in cancer-related genes. Based on these alterations, they suggest potential targeted therapies.
• Limitation: These tests predict drug responses based on known associations between genetic changes and drug sensitivity. However, not all genetic changes directly result in drug sensitivity, leading to potential false positives. Additionally, many tumors have no known actionable mutations.
• Protein Expression Assays (e.g., IHC - Immunohistochemistry)
• Application: IHC assesses the presence of certain proteins in tumor samples. For instance, HER2 expression in breast cancer predicts response to drugs like trastuzumab.
• Limitation: IHC is usually specific to one or a few proteins. It doesn't provide a comprehensive overview and can miss potential drug targets. Moreover, protein levels don't always correlate with activity or drug sensitivity.
• Functional Assays (e.g., ChemoFx, Oncotest)
• Application: These assays expose tumor samples to various drugs to observe direct drug effects on tumor cells.
• Limitation: Such tests can be time-consuming, costly, and might not always reflect in-vivo responses. Some also require viable tumor tissue, which might not always be available.
• Phenotypic Screens
• Application: High-throughput screens expose cancer cells to many drugs to observe effects.
• Limitation: While comprehensive, these screens can be costly and time-consuming. Moreover, they assess drug response in an artificial in-vitro environment, which might not always correlate with in-vivo responses.

The potential advantages of Function's solution include:

• Direct Functional Insight: Unlike genomic profiling that identifies mutations without confirming their impact on drug sensitivity, the CRISPR platform directly assesses gene function, providing a more reliable prediction of drug response.
• Unbiased Comprehensive Assessment: The platform doesn't rely on prior knowledge of gene-drug associations, making it capable of uncovering novel drug sensitivities.
• Dynamic Monitoring: The platform can potentially be used at multiple timepoints, tracking changes in drug sensitivity as the tumor evolves, enabling adaptive treatments.

Commercial opportunity: predicting response to drugs

Tools that predict drug-tumor response play an increasingly vital role in precision oncology, but their commercial model faces multifaceted challenges and opportunities.

Reimbursement for diagnostic tests or tools in oncology, especially those involving genomics or advanced technologies, can be challenging. Insurers, including government programs like Medicare, are typically conservative in adopting new technologies. They need substantial evidence of clinical utility. The reimbursement landscape can vary by region, payer, and even from one patient population to another. This can make it challenging for providers to know whether they'll be reimbursed for a given test or tool.

To secure reimbursement, companies must demonstrate the clinical utility of their tool, meaning they have to show that the test provides information that can guide and improve patient outcomes. This includes generating evidence from well-designed clinical trials that show patients managed with the tool have better outcomes (e.g., response rate, progression-free survival) than those managed without it. In addition to clinical trials, payers often look at real-world evidence, including case studies, retrospective analyses, and observational studies that show the tool's effectiveness in a "real-world" clinical setting.

Tools must also be cost-effective, demonstrating that the tool, even if it is expensive, can lead to overall cost savings by ensuring patients receive the most effective treatment up front, thereby potentially avoiding costlier interventions later.

These studies can be costly and time consuming, and there is some risk that the study's results do not support use of the test.

Pricing strategies for these tools can vary widely based on several factors:

• Competitive Landscape: If several similar tools are on the market, there may be competitive pricing pressure.
• Innovation and Uniqueness: Highly innovative or unique tools that offer significant advantages over existing solutions can command a premium price.
• Comprehensive vs. Single-Gene Tests: Comprehensive genomic panels that test multiple genes simultaneously are generally more expensive than single-gene tests.

Examples of tools in this space include the Oncotype DX for certain breast cancers, which helps predict the benefit of chemotherapy and the risk of recurrence, and FoundationOne CDx, a comprehensive genomic profiling test for all solid tumors. Their pricing varies, with comprehensive tests typically being more expensive, often reaching several thousand dollars. However, the exact price can depend on negotiated rates with payers and other factors. Guardant Health expects to generate $550M in 2023 revenue, of which Guardant360 is a large part.

Reimbursement

Securing reimbursement from Medicare for a diagnostic test involves demonstrating both clinical utility and clinical validity.

Here's a breakdown of the studies and data that Function Oncology might need to consider for Medicare reimbursement for its Fx Heme test (this could apply to other cancers as well):

• Prospective Clinical Trials: While the retrospective study demonstrates promise, prospective studies in which patients are treated based on the test's recommendations would provide stronger evidence. This study should ideally:
• Demonstrate improved patient outcomes when treatment decisions are made based on the Fx Heme test compared to standard-of-care treatment decisions.
• Indicate an increased likelihood of treatment success, decreased unnecessary treatments, or better patient survival rates.
• Head-to-Head Comparison with Existing Tests: Comparing the Fx Heme test's efficacy with currently reimbursed tests would provide a clear advantage. This will help establish:
• Superior accuracy and sensitivity in predicting drug responses.
• Clinical advantages over other testing methodologies, like shorter turnaround times or better actionable insights.
• Cost-Effectiveness Analysis: Demonstrating the economic value of the test is crucial. By showcasing that early and accurate detection could lead to cost savings in the long run (e.g., by reducing ineffective treatments), Medicare would be more incentivized to provide coverage.
• Real-world Evidence (RWE) and Post-market Surveillance: Gathering real-world data, including patient outcomes, cost savings, and patient or physician satisfaction, can further support the test's clinical utility.
• Pivotal Validation Study: An extensive multi-center study confirming the clinical validity of the test – its sensitivity, specificity, positive predictive value, and negative predictive value in a larger, diverse patient population.
• Analytical Validity: Demonstrating the reliability and reproducibility of the test results across various labs, ensuring consistency in patient testing no matter where the test is administered.
• Utility Studies: Beyond just predicting treatment outcomes, it would be beneficial to show the test's broader impacts. For instance, can it guide earlier interventions? Can it reduce the use of ineffective treatments, hence leading to cost savings and improved patient quality of life?
• Stakeholder Engagement: It's also essential to engage stakeholders, such as oncologists, patients, and patient advocacy groups. Their input and testimonials regarding the value of the test can be instrumental.

Tests like FoundationOne CDx and Guardant360 are used to guide treatment decisions in oncology, and their reimbursement challenges can offer lessons. Clinical utility – proving that the test results in better patient outcomes – has historically been a sticking point for many diagnostics. Both of these tests have secured reimbursement after extensive data collection and demonstrations of clinical utility.

Additionally, with the rise of precision medicine, the field has been moving towards "value-based" reimbursement. This might mean that tests like Fx Heme need to provide ongoing evidence of their value to continue securing reimbursement.

Both FoundationOne CDx and Guardant360 faced challenges in securing reimbursement, but their successful navigation of this process provides valuable insights.

• FoundationOne CDx
• Clinical Validity and Utility Studies: FoundationOne CDx conducted comprehensive genomic profiling (CGP) studies that demonstrated the ability of their test to identify actionable mutations and guide therapy, as well as its potential to match patients to clinical trial opportunities.
• Validation Study: Initially, FoundationOne (the precursor to FoundationOne CDx) was subjected to a validation study involving more than 2,000 samples, showing a high concordance rate with traditional methods.
• Clinical Utility Studies: These studies typically focused on the test's ability to detect actionable mutations that can guide therapy decisions. For instance, one study might involve hundreds of patients across multiple cancer types and would measure endpoints like the percentage of patients for whom an actionable mutation was detected and subsequently treated based on the test's findings.
• Retrospective Analyses: Using its vast database, Foundation Medicine conducted retrospective analyses to correlate genomic findings with clinical outcomes, helping to further prove the test's utility.
• FDA Approval Study: The approval of FoundationOne CDx as a companion diagnostic was backed by a clinical validation in which over 300 samples were compared to FDA-approved tests, showing a very high concordance.
• FDA Approval: FoundationOne CDx went the extra mile by securing FDA-approval as a companion diagnostic for specific cancer types and therapies. This not only added credibility but also provided a strong case for its clinical validity.
• Collaborative Agreements with Payers: Foundation Medicine, the company behind FoundationOne CDx, entered into collaborative agreements with major insurance providers to cover the test. These collaborations often involved sharing of data and outcomes to continuously validate the test's utility.
• Real-world Evidence: Foundation Medicine accumulated a vast database of test results and associated clinical outcomes, which served as ongoing evidence of the test's utility in the real-world setting.
• Guardant360
• Clinical Validation Trials: Guardant360 conducted numerous studies demonstrating the test's high sensitivity and specificity in detecting actionable mutations from circulating tumor DNA (ctDNA) in blood samples. These trials typically involved hundreds of patients and compared Guardant360's liquid biopsy results against results from tissue biopsies to determine sensitivity, specificity, and overall concordance. For instance, one trial might involve a few hundred patients with non-small cell lung cancer (NSCLC) and measure how often Guardant360's findings matched those of a traditional tissue biopsy.
• Prospective Clinical Trials: The company ran prospective trials to show that using the test led to actionable findings that could guide therapy, and in some cases, these findings were associated with improved patient outcomes. These trials were designed to assess clinical utility. A key endpoint in such trials might be the "actionability" of the mutations detected by Guardant360. That is, what percentage of patients had a mutation detected that could directly guide therapy decisions?
• Comparison with Tissue Biopsies: Guardant360 presented data comparing its liquid biopsy approach with traditional tissue biopsies, showing comparable efficacy and, in some cases, advantages such as faster turnaround times and less invasiveness.
• Collaborations with Pharmaceutical Companies: Guardant360 collaborated with pharma companies for use as a companion diagnostic, strengthening its position in the market and its argument for clinical utility.
• Data on Cost-Effectiveness: Guardant Health, the company behind Guardant360, also presented data on the potential cost savings associated with early and accurate detection using its test, making an economic argument in favor of reimbursement. While not a clinical study, these analyses were important for reimbursement considerations. For instance, Guardant Health might analyze the potential cost savings associated with avoiding a traditional tissue biopsy in favor of a Guardant360 test.

Improving clinical trial design

Patient stratification for clinical trials is a method of categorizing patients into subgroups based on specific criteria or biomarkers. This approach aims to identify the individuals most likely to benefit from a particular treatment. By using a technology like Function Oncology's CRISPR-enabled functional genomics profiling platform, patient stratification can be greatly refined.

• Enhancing Clinical Trial Efficacy
• Targeted Enrollment: By understanding individual gene dependencies and vulnerabilities to drug targets, researchers can enroll patients who have a higher likelihood of responding to the investigational drug. This not only increases the chances of trial success but also can lead to clearer, more interpretable results.
• Reduced "Noise" in Data:: In traditional clinical trials, where patient populations are more heterogeneous, non-responders or adverse reactions from a subset of patients can mask the benefits seen in another subset. Stratification reduces this variability, allowing for clearer identification of drug effects.
• Accelerated Drug Approval
• Efficient Trials: Stratified clinical trials might require fewer participants since the probability of observing a significant therapeutic effect is higher. This can expedite the trial process.
• Regulatory Preference: Regulatory agencies like the FDA often favor treatments that have a clear identified target population, especially if it's backed by solid genomic or molecular evidence. This can lead to faster drug approvals.
• Economic Implications
• Cost Savings: Clinical trials are one of the most expensive stages in drug development. Improving their success rate and potentially reducing the number of participants can lead to significant cost savings for pharmaceutical companies.
• Return on Investment: Drugs that have a clear target demographic and high efficacy within that group can command premium pricing and might see faster uptake in the market.
• Enhanced Safety Profile
• Minimizing Adverse Events: By understanding the genetic makeup of participants and their potential vulnerabilities, it's possible to anticipate and minimize adverse reactions.
• Ethical Considerations: It's ethically favorable to provide a treatment to those who are more likely to benefit from it, reducing potential harm or unnecessary side effects to non-responders.
• Facilitating Personalized Medicine
• Biomarker Discovery: During the stratification process, researchers might discover new biomarkers that predict drug response. These biomarkers can then be used in clinical settings to identify patients who might benefit from the drug once it's approved.
• Tailored Therapies: Clinical trials using stratified cohorts can lead to the development of treatments that are tailor-made for specific patient groups, ushering in an era of truly personalized medicine.
• Strengthening Trust and Enthusiasm in Clinical Trials
• Patient Confidence: When patients understand that they've been selected for a trial based on their unique genetic makeup and potential to benefit, it can increase their confidence and enthusiasm for participation.
• Physician Engagement: Physicians might be more inclined to refer their patients to trials if they believe that the trial is more specifically suited to their patient's genetic or molecular profile.

Patient stratification in clinical trials represents a significant market opportunity within the drug development industry. As pharmaceutical companies seek to improve the efficiency and success rates of their clinical trials, providers that can offer effective stratification solutions will be well-positioned to capture a growing market segment.

Pharma companies often form partnerships with biotech firms or service providers like Function to leverage their advanced technologies for patient stratification. These partnerships can be structured as licensing agreements, where the pharma company gets access to the technology, or as collaborative research agreements.

Given the high-risk nature of drug development, many agreements are milestone-based. Providers may receive payments upon achieving specific milestones, such as successful patient identification, progression to the next phase of trials, or final drug approval. In some cases, large pharma companies might take an equity stake in smaller biotech firms, effectively betting on the success of their stratification technology and the associated drug pipeline.

Global pharmaceutical R&D expenditure was continuously rising, with estimates putting it at over $180 billion annually. Much of this is directed at oncology. With increasing interest in personalized medicine and targeted therapies, a substantial portion of this amount will likely be directed towards technologies facilitating patient stratification.

The number of registered clinical trials has been increasing year-over-year, with thousands of new trials initiated annually. If even a fraction of these trials adopt advanced stratification techniques, the market potential is vast.

A majority of clinical trials do not reach the market, often due to lack of efficacy. Improved patient stratification can enhance the success rate, making investment in this area a priority for many pharmaceutical companies.

There are several companies that have successfully built products and services around improved patient stratification for oncology trials. Some of these companies have developed platforms that leverage genomic, transcriptomic, proteomic, or other omic data to identify patient subsets most likely to benefit from specific therapies:

• Foundation Medicine
• Product/Service: Comprehensive genomic profiling tests that identify the molecular alterations in a patient's cancer and match them with relevant targeted therapies, immunotherapies, and clinical trials.
• Significance: Foundation Medicine's tests, such as FoundationOne®CDx, have become integral in oncology practice for tumor profiling and guiding treatment decisions based on genomic alterations.
• Tempus
• Product/Service: Uses artificial intelligence, machine learning, and genomic sequencing to better understand a patient's tumor, thus aiding in personalized cancer care and research.
• Significance: Tempus has grown rapidly and collaborates with various academic medical centers, NCI-designated cancer centers, and pharmaceutical companies to refine treatment approaches based on patient molecular profiles.
• Guardant Health
• Product/Service: Liquid biopsy tests, like Guardant360, that detect tumor DNA circulating in the blood, providing genomic information to guide treatment decisions without the need for invasive tissue biopsies.
• Significance: Guardant's tests have been critical in situations where tissue biopsies are not feasible or too risky.
• Invitae
• Product/Service: Offers a range of genetic tests, including those that can identify inherited risks for certain cancers, helping in early detection, prevention, and tailored treatment.
• Significance: Invitae has been actively involved in genetic risk assessment and provides insights that can guide both treatment and preventive strategies in oncology.
• Caris Life Sciences
• Product/Service: The company provides molecular profiling solutions, including Caris Molecular Intelligence, which offers genomic, transcriptomic, and proteomic insights to guide more personalized cancer treatments.
• Significance: Caris has been a leader in multi-omic tumor profiling, aiding oncologists in understanding the unique characteristics of a patient's tumor.
• Personal Genome Diagnostics (PGDx)
• Product/Service: PGDx provides tissue-based and liquid biopsy genomic profiling tests to identify alterations in cancer-related genes.
• Significance: PGDx's solutions enable better stratification of patients for clinical trials and help guide therapeutic decisions in clinical practice.
• NeoGenomics
• Product/Service: Provides a suite of oncology testing services, including molecular genetic testing, that assists in diagnosis, prognosis, and treatment decisions for cancer patients.
• Significance: As one of the leading oncology reference laboratories, NeoGenomics plays a pivotal role in patient stratification for clinical trials and routine care.

Drug repurposing

Function Oncology's CRISPR-enabled functional genomics profiling platform offers a promising avenue for drug repurposing or expanding the use of existing cancer drugs.

Utilizing CRISPR/Cas9 technology, this platform offers an unbiased evaluation of gene dependencies across various cancer types. Instead of just focusing on mutations traditionally associated with specific cancers, this platform can shed light on a broader array of gene dependencies, potentially uncovering new targets for existing treatments.

As the platform identifies gene dependencies not traditionally linked with specific cancers, there's a potential to discover new applications for existing drugs. If a gene, traditionally associated with one cancer type, emerges as a dependency in another, existing treatments might find new applicability.

The platform's ability to reveal cases with dependencies on multiple drug targets offers an opportunity to strategize combination therapies. Addressing multiple dependencies concurrently might enhance therapeutic efficacy.

The platform can be used to help pharma companies identify repurposing opportunities and to enable Function to develop its own repurposed therapies.

Developing novel cancer therapeutics

Function's platform offers a promising approach to identifying gene dependencies in acute myeloid leukemia (AML) cells. By comprehensively assessing the effects of knocking out genes corresponding to all FDA-approved oncology drug targets, the platform seeks to identify genes that, when inhibited, can have a therapeutic effect on AML (as well as other cancers).

• Advantages
• Precision and Specificity: CRISPR/Cas9 is a precise genome-editing tool, allowing for targeted knockouts of specific genes. This specificity is crucial for establishing clear cause-and-effect relationships between gene function and cellular phenotype.
• Comprehensive Assessment: By targeting the genes corresponding to all FDA-approved oncology drugs, the platform can identify potential repurposing opportunities for existing drugs and provide insights into combination therapies.
• Functional Relevance: Unlike pure genomic profiling, which provides a static snapshot of gene mutations and expression, functional genomics assesses the actual impact of gene knockouts on cell behavior, enhancing the likelihood of discovering clinically relevant targets.
• Disadvantages
• Complexity of Disease: Cancers, including AML, are heterogeneous and can evolve over time. A target that's relevant in one patient or at one disease stage may not be relevant in others.
• Validation: Discovering a gene dependency in vitro doesn't necessarily translate to effective drug targets in vivo. The biological context of a living organism can affect drug efficacy and toxicity.
• Drug Development: Even if a novel target is identified, the journey from target identification to a marketable drug is long, expensive, and fraught with potential failures.

Genomic studies, such as The Cancer Genome Atlas (TCGA), have identified numerous genetic alterations in various cancers. However, only a fraction of these have translated into actionable drug targets. Genomic information alone sometimes lacks the functional context necessary to determine which mutations are drivers of the disease and which are mere passengers.

Techniques like RNA interference (RNAi) have previously been used for large-scale functional screens. While they've identified potential targets, RNAi has limitations, such as off-target effects, that can complicate results. CRISPR/Cas9 offers a more specific and efficient alternative, but it too is not without challenges, including potential off-target edits and cellular compensatory mechanisms.

Some large-scale genomic and functional platforms have yielded actionable insights and drug candidates. For instance, RNAi screens have contributed to the understanding of essential genes in cancer cells. However, many identified targets have not yet resulted in FDA-approved drugs, highlighting the complexity of translating basic research into clinical applications.