Advent of Ex Vivo testingThe last two decades have seen dramatic developments in gene sequencing. Spurred by financial incentives, the field developed rapidly and even beat Moore's law of electronics. Currently, the genome of can be analyzed with around $1000. Nevertheless, there are limited applications of genomics for clinical or therapeutic purposes.
A recent analysis from 2006 to 2018 was performed on 31 drugs with 38 FDA-approved indications that met inclusion criteria for genome-targeted or genome-informed therapy. Percentages of patients eligible for the therapies are show in Table 1.
Table 1: The percentages of patients who were eligible for therapies in 2006 and 2018
The summary above suggests that eligibility is low, or 1 in 12 and 1 in 6 for genome-targeted and-informed therapies, respectively, albeit increased by 50% to 60% between 2006 and 2018. Moreover, the median overall response rate for all genome-informed drugs through January 2018 was 54%, meaning that the minority available for such genomic therapy had chances of benefiting similar to a coin toss.
Our Ability to Analyze Genomic Information is Limited
One interpretation is that our ability to analyze genomic information about patients is incomplete, and effectiveness will increase with more sequencing and deeper understanding of interconnections. However, biology presents some of the most complex systems, and their breakdown involves understanding not only of up to 20,000 genes, but also interactions between different cells types, the extracellular matrix, the immune system, diet and metabolism, as well as a host of other factors. Predicting the response of such complex systems based on analysis of a few genes has severe limitations. Instead of analyzing individual genes or biomarkers, a more powerful and holistic test is to use the biology itself for the computational purposes.
The Importance of Bio-Computation
This concept is called bio-computation because it has a strong emphasis on using biology for "computational" purposes. Due to concepts like "emergence" which means that the total is greater than the sum of the parts, there are great limitations for the application of AI to biology. For example, "wet computation" by living biology has a greater ability to capture complexity, while AI operates on a handful of genes and is more likely to miss emergent of results. In other words, the probability of positive responses to a given treatment is conditional. Thus, each piece of information that AI misses can substantially reduce the predictive potential of the treatment.
Shortcomings of Oncology Treatments
Most of us have a friend or a relative who had been diagnosed with cancer in which case treatment is personal. The reality is the treatment will either work or will not. Nevertheless, relying on statistics is not good enough. Many oncology treatments that do not use biomarkers have a 30-50% response rate. In the end, if a treatment does not work for you or a loved one, it does not matter if it eventuality had either a 1% or a greater than 50% chance of producing good results. We all expect is doctors to give their best effort to achieve good outcomes. Technological shortcomings with biomarkers or full diligence on predicting health outcomes are irreverent of our grief.
Triple-Negative Breast Cancer
A stark illustration of this is triple-negative breast cancer (TNBC) -- breast cancer not tied (does not over-express the genes) to any of the three likely culprit receptors: estrogen, progesterone, and HER2/neu. Being unsuitable for hormone therapies targeting one of the three receptors, TNBC patients often require combination therapies. Recently, pharmaceutical companies and hospitals have begun testing combinations of approved drugs. In principle, the combinations with the abilities to capture the essence of pathology and respond to a treatment, called biomarkers, can be used to guide patients to the best treatments.
Biomarkers are being co-developed with therapies from the early preclinical stages in recognition of their effectiveness to improve drug development and gain regulatory approval by showing high efficacy in biomarker-targeted therapy. However, these studies are both expensive and time consuming: while over a third of new drug development has associated biomarker programs, this is not the case for repurposing established drugs. The economic case has significantly slowed adoption.
There is a new technique that is commercially available -- ex vivo testing. In the most common application, a needle biopsy of a solid tumor is used to provide source tissue which can be tested outside of the body. This technique applies to any physiologically relevant test, including those for neurodegenerative and rare diseases. One of the key technologies for this purpose is micro-physiological systems (MPS). In MPS, there is typically solid, 3D tissue at the core of such platforms, but there are other elements such as white blood cells for the treatment of immuno-oncology or for capturing and cultivating circulating tumor cells (CTCs) which are shed by solid tumors to provide early warning signs of metastatic potential with a blood draw, which is less invasive than other techniques.
FlowCell is partnering the companies like Biolidics to provide access to liquid biopsy tools. Our CSO Lahiri Nanduri has done extensive testing and validation of their technology.
We are also partnering with other organizations that provide leadership in the field of precision oncology ex vivo testing: Kiyatec, SageMedic, Lena Biosciences, Biomarker Strategies, and others. These life-saving technologies are complex as are the business executions. Many organizations are shifting from providing tools to providing services, while other leaders in the field, such as Kiyatec, are fully commited to this business model.
Per requirements of specific applications, the field of test platforms, miniaturization for human physiology outside of the body, is fragmented. The consortium of innovators (and enlightened customers), NA3RsC, is focused on advancement in MPS technologies, with technical standardization and outreach to regulators like the FDA, and international consortiums with similar goals like IQ MPS. The organization is aligning the stakeholders, with dedication to Reduce, Refine, and Replace animal models (3Rs) with MPS platforms. There is regulatory pressure: by 2025 animal models are expected to be reduced by 30%, and eliminated entirely by 2035, except for special cases. Companies like Loreal are already working with MPS models of skin, after animal testing of skincare products was prohibited in EU in 2012. The implications for the much larger pharmaceutical industry are profound. Some small innovators have based their busines on MPS platforms, like Kiyatec, Mitra, and Nilogen, but for larger companies it has largely been a series of one-off projects. This is due to lack of standardization especially in validation of MPS platforms and regulatory guidance.
The current impasse is due primarily to lack of education about existing capabilities, and lack of business model and under-investment. Specifically, while the pharma industry embraced and subsidized the genomic revolution, the micro-physiological revolution which has dramatic potential for both cost-savings and patient outcomes has seen relatively little funding. That is a reflection of the relative simplicity of ACTG basepair sequencing versus recapitulation of physiological conditions for a biological model with strong fidelity or predictive potential of patient response.
Ex vivo applications in a traditional drug development paradigm are becoming intricate parts of the preclinical drug development phase. [Figure 1] Preclinical ex vivo testing limits its potential, while application in clinical trials has been very slow due to technical complexity and economic reasons -- reluctance to limit the patient base with additional information.
Figure 1: Ex vivo applications in traditional drug development are becoming intricate parts of the preclinical drug development process
There is a lack of investment while pharma and regulators await maturation of MPS technologies. One of the ways this conundrum can be addressed is through community or democracy.
Our solution involves bypassing regulators and the drug developers with lengthy clinical trials and focusing of individualized medicine. We contact hospitals and patients directly to create awareness of ex vivo testing tools can save lives: #SaveGrandma is not just about COVID, masks, and social distancing, but affects other pathology such as cancer. At this time, the major barriers for individualized medicine are lack of awareness of available technologies and services, as well as financial reimbursement. #SaveGrandma may require GoFundMe campaign to finance precision oncology treatment.
Our goal is not to leave this burden to the families and friends of a cancer patient, but to lobby for insurance reimbursement for such life-saving treatments. The services section of our website has information for several providers of precision oncology and other services (neurodegeneration, toxicology, etc). We are creating a website Health-Sherlock for empowering patients and democratization of healthcare data for collection of the largest dataset, which will be especially helpful to patients with rare diseases, including many cancer subtypes. We are also raising funds to empower this development and hire some dedicated team-members full time for web development, contract- and technical writing.
Please contact us if you know someone with a challenging cancer case, or a rare disease in need of differential diagnosis and effective treatment of the cause, not just symptoms. We can offer the former practical testing options to find efficacious treatment. For the latter we will provide access to joint experience extracted from a repository of de-identified patient data, as well as tools to analyze clinical trials (Natural Language Processing) and scientific publication to keep patients and healthcare provides aware of the lasts developments in their field.