Abstract
As precision medicine has become increasingly important in cancer therapy, researchers have grappled with how to implement it in the clinic. Three studies demonstrate how different approaches to precision medicine—molecular profiling using individually customized combination therapy, RNA transcriptomics, and circulating tumor DNA sequencing—can be used to match patients to targeted treatments and trials.
Rapid advances in genomic sequencing and cancer therapeutics have led to an increased focus on precision medicine, yet this personalized approach to treatment has yet to be routinely incorporated into patient care. Three recent studies demonstrate how different strategies—molecular profiling, RNA transcriptomics, and circulating tumor DNA (ctDNA) sequencing—can be used to match patients to targeted treatments and trials. These studies, though preliminary, provide proof of principle that such strategies can be deployed in the clinic.
Razelle Kurzrock, MD, of the University of California, San Diego, and colleagues developed the I-PREDICT trial because the standard strategy of matching patients to monotherapies based on single mutations may not be effective for those with multiple alterations. “We liken it to malignant snowflakes—each patient with a metastatic tumor has a complicated genomic sequence, and has a different sequence,” Kurzrock says.
In I-PREDICT, the researchers performed next-generation sequencing on tumors from patients with relapsed/refractory metastatic solid cancers and used the results to match patients to customized drug combinations, some of which had not been previously tested in clinical trials. (In these cases, drug doses were carefully titrated to ensure safety.) Of 149 patients initially enrolled, 49% received matched therapies, which were often combinations of drugs rather than single drugs, far exceeding the 5% to 10% rate seen in many previous monotherapy trials—a difference likely due to factors such as broader testing panels, timelier recommendations for specific therapies, and greater drug availability.
Patients matched based on more of their total alterations fared better than those matched based on fewer of their total alterations—patients with a high matching score had a disease control rate of 50% and a median progression-free survival of 6.5 months, and did not reach median overall survival, compared with 22.4%, 3.1 months, and 10.2 months, respectively, in patients with a low matching score.
Kurzrock considers the trial proof that patients can be safely and effectively matched to customized drug combinations based on multiple alterations, and that matching patients based on more alterations leads to better outcomes. She and her team are now testing this approach in newly diagnosed patients.
Kurzrock also co-led the WINTHER trial, an international study from the WIN consortium for personalized cancer therapy. WINTHER investigated the utility of RNA transcriptomics for matching patients with relapsed/refractory solid cancers to therapies. The trial included 69 patients who were matched to and received therapies based on next-generation DNA-sequencing results, and 38 who received matched therapies based on RNA expression; patients were preferentially matched based on DNA sequencing, moving to the RNA arm only if they couldn't be matched based on the DNA. In the DNA arm, 23.2% of patients had a partial or complete response to their therapy and stable disease for at least 6 months, compared with 31.6% on the RNA arm, a difference that was not statistically significant. Adding RNA expression to DNA sequencing increased the rate of patients who matched to and received therapies from 23% to 35%.
“The clinical takeaway is that RNA transcriptomics should be added to genomics—it gives us an important extra component beyond what DNA gives us,” Kurzrock says. She and her WIN consortium colleagues recently launched the SPRING trial, which integrates DNA sequencing and RNA transcriptomics to better understand which patients respond to triple-drug combinations.
The TARGET study assessed another potential tool for precision medicine: ctDNA. “We set out [asking], ‘Can we utilize next-generation sequencing results from ctDNA to match patients to phase I trials?’” explains Matthew Krebs, MB, ChB, PhD, of the University of Manchester (UoM) in the UK, who co-led the study with Caroline Dive, PhD, of the Cancer Research UK Manchester Institute, UoM.
The researchers performed next-generation sequencing on ctDNA and tumor samples from 100 patients with advanced cancers. They found that the results aligned and could be obtained within an average of 33 days (in the initial feasibility phase of the trial) and 30 days, respectively. Moreover, ctDNA and tumor profiling identified actionable mutations in 41 patients, 11 of whom received a matched therapy on a phase I trial; four of the 11 responded to therapy.
“From the data we have, we think doing a blood sample and tumor analysis are complementary to one another and give the best chance to finding an actionable alteration to try to match patients to trials,” Krebs says, noting that although the match rate was low—and hopefully can be improved—the trial demonstrates the feasibility of the process and the culture change in industry to permit ctDNA results for inclusion of patients in their studies.
“What these three papers show is that precision medicine in the setting of a phase I trial is actually a very viable and sensible option,” Dive adds. “There's a lot of improving to do, but all three papers are pointing us in the right direction.”
For Vivek Subbiah, MD, of The University of Texas MD Anderson Cancer Center in Houston, who wasn't involved in the studies, they demonstrate the value of moving away from the traditional clinical trial paradigm: “We need to have novel, out-of-the-box clinical trial ideas,” he says. “These studies are, I think, first steps in our understanding towards real-world human data—they can propel the field of precision oncology forward.” –Catherine Caruso