Triomics Secures 15M USD to Automate Cancer Trial Matching

Triomics secures 15M USD to invest in their startup to create an AI that will help speed up the process of matching cancer clinical trial patients.

Therapeutic interventions implemented in clinical trials can potentially prolong cancer patients’ lives. 

However, despite the annual conductor of thousands of clinical trials in the United States, enrollment in investigations of novel treatments remains between 3% and 5% of eligible patients.

The generative AI startup Triomics asserts that it can substantially decrease the time required for physicians to connect patients with clinical trials.   

Recommendation letters from physicians are often crucial in enrolling patients. However, oncologists and nurses who are frequently preoccupied with their patients’ care lack the time to research every possible clinical trial.  

Because I am not a physician, I am unaware of the daily obstacles that oncology medical staff face. However, I can attest from personal experience to the difficulty of locating clinical trials for cancer patients. I devoted countless hours to perusing clinicaltrials.gov, a website, and database that catalogs thousands of ongoing trials when my father was ill. Furthermore, I devoted a portion of a Saturday in March to locating a clinical trial for an acquaintance diagnosed with stage IV cancer. Since her physician had only provided a single trial, she inquired whether alternative courses of action were available.  

Eligibility requirements for most clinical trials are intricate, encompassing dozens of factors such as cancer stage, mutations, and prior treatments. Manually determining an appropriate clinical trial from a patient’s medical record frequently requires hours of medical staff time. However, a shortage of oncology specialists prevents many cancer patients from being invited to participate or causes them to fail the eligibility deadline. 

Former MIT biotechnology researcher Sarim Khan and Adobe AI scientist Hrituraj Singh established Triomics. Having been friends since college, the two individuals resolved to found Triomics in 2021 after realizing that developments in generative AI and LLMs could assist in the extraction of data from electronic health records (EHR) to locate suitable clinical trials for cancer patients in minutes rather than hours.

In the winter of 2021, Khan and Singh joined Y Combinator and commenced the development of an LLM tailored for hospital systems’ oncology departments and cancer facilities.  

Triomics intends to double the number of cancer centers and hospitals actively utilizing or piloting its LLM by the end of the year, three years later. Currently, six such facilities are on board. The company has secured $15 million in Series A funding from Lightspeed, Nexus Venture Partners, General Catalyst, and Y Combinator to continue developing and distributing its platform to new consumers. 

Although the most apparent benefit of Triomics software may be reducing the time required to match patients with clinical trials, Khan explains that Triomics is much more than a clinical trials organization. “Doctors employ it for a wide variety of use cases, the list of which I could go on and on,” he explained. 

Triomics’ LLM, which the company calls OncoLLM, “reads” the patient’s medical record. The data could assist medical staff in preparing for patient visits or in submitting cancer data to state regulatory agencies, including information about affected organs and progression stages.  

Certainly, Triomics is not the only company addressing this matter. Additional firms engaged in matching AI clinical trials consist of Deep 6 AI, QuantHealth, and Trajectory, among others. 

However, Khan thinks that Triomics is among the few startups that exclusively handle extensive datasets for cancer centers.