Q: Clinical Trials is such an important topic, but it is so broad. Is there a way to provide better matching for Clinical Trials involving cancer patients, especially those whose cancers may have the added diagnostic information of molecular profiling?
A: Clinical trials provide cancer patients, especially those whose cancer has progressed beyond standard of care, access to potentially promising investigational therapies or novel combinations of approved therapies. Yet only a small fraction (less than 5%) of the 1,700,000 Americans diagnosed annually with potentially lethal cancer participate in clinical trials. A growing number of cancer patients now have their tumors profiled molecularly and there is real opportunity to use that information in better matching them to clinical trials. Below I outline the CollabRx approach to this problem.
Most clinical trials for cancer seek patients based on criteria such as tissue of origin and/or histological characteristics of the tumors. But with hundreds of targeted therapies now approved or under development, there are a growing number of clinical trials looking for patients based on the molecular characteristics of their cancer. Of note are genomically-driven trials such as basket and umbrella trials that match patients to therapies based on the individual molecular profiles of their individual cancers.
While such trials are a natural fit for patients whose tumors have been molecularly characterized, they are not easy to identify in clinicaltrials.gov. For example, a search for the gene ‘BRAF’ identifies trials for patients with BRAF mutations as well as those without. This issue can be resolved only with careful annotation and structuring of trial inclusion and exclusion criteria from sources.
Some trials without explicit molecular criteria may ALSO be relevant to patients with certain molecular profiles. What are these trials and how can physicians and patients identify them?
A patient whose cancer has a certain molecular profile (defined as having a particular set of variants) may benefit from participating in a clinical trial if the trial therapy can be “associated” with one or more individual variants in the patient’s molecular profile. For example, if:
- The therapy either directly mitigates the oncogenic effect of one or more variants in the patient’s tumor or a downstream component of a relevant biological pathway
- There are published data suggesting that one or more variants are predictive of response to therapy
- Clinical experience of oncologists/cancer centers call for use of therapy for a patient with a certain molecular profile
At CollabRx, we use ‘treatment strategies’ to match cancer patients with trials. Treatment strategies explicitly connect a drug or drug class to a variant or variant class and a diagnosis. For example, a treatment strategy for ‘BRAF activating mutations’ in thyroid cancer could include ‘BRAF inhibitors’ (drugs that directly target BRAF) and, say, ERK inhibitors (if published data were to suggest efficacy of ERK inhibitors in BRAF-mutated thyroid cancer). We then match any trials evaluating BRAF inhibitors or ERK inhibitors to patients with BRAF activating mutations, even if those trials don’t have “BRAF activating” molecular criteria.
We believe that matching patients to clinical trials using this two-pronged approach–using molecular inclusion criteria AND using strategically relevant therapies–could identify treatment options for patients that they may not have considered otherwise.