Analyzing tumors for DNA mutations and RNA expression significantly increased the number of patients with cancers that could be matched to recommended therapies, investigators in an international precision oncology initiative reported.
Overall, 35% of patients had tumors that could be treated with precision therapies recommended by a panel of experts. Previous studies limited to DNA profiling identified recommended therapies for 5%-25% of patients. However, the trial did not meet the primary endpoint of achieving progression-free survival (PFS) with recommended therapy that exceeded the PFS achieved with prior therapy by more than 50%.
The results, reported in Nature Medicine, set the stage for several ongoing trials evaluating simultaneous evaluation of DNA mutations and RNA expression profiles in patients with different types of cancer.
“It’s really clear that DNA is just the tip of the iceberg,” said coauthor Razelle Kurzrock, MD, of the Moores Cancer Center at the University of California San Diego. “We learn a lot from DNA but we don’t learn everything from DNA. We match patients on the basis of DNA and we don’t always get the results we want. There are a lot of potential reasons for that, but one of the reasons is that DNA is not giving us the full picture.”
Tumor DNA analysis has provided the cornerstone for the emerging field of precision oncology: identifying genetic aberrations in tumors and then treating the tumors with therapies that target the abnormalities. Reflecting the complexity of cancer, only a small minority of tumors have DNA mutations that can be matched with an appropriate targeted therapy. RNA transcription analysis (transcriptomics) takes evaluation of tumor genetics to deeper levels of the underlying biological mechanisms.
Investigators at cancer centers in Israel, France, the U.S., and Canada participated in the WINTHER trial to test the hypothesis that examining both DNA mutations and RNA transcription would increase the number of patients with targetable cancers. The study included 303 patients with various types of advanced cancer (colon, lung, and head and neck accounting for 72% of the total). The cohort had a median treatment history of three prior regimens. The primary endpoint was greater than 50% improvement in PFS with the recommended therapy versus prior therapy (PFS2/PFS1 ratio >1.5) in at least 50% of patients.
Using tumor biopsy specimens, investigators initially performed DNA sequencing studies with a 236-gene panel, which identified 69 patients with drug-matched alterations. The tumor tissue was analyzed a second time for RNA expression (comparing tumor and normal tissue), which identified an additional 38 patients with tumors that could be matched to drugs targeting specific abnormalities. The two rounds of analysis resulted in a 35% rate of genetics-drug matching.
Matched therapies were not limited to targeted agents. The 107 patients received a total of 159 drugs (monotherapy and combinations), which included targeted therapies, immunotherapies, and hormonal therapies. Treatments included approved (n=22), approved but off label (n=115), and investigational therapies (n=22).
The 107 patients treated with matched therapies had a clinical benefit rate (response plus stable disease) of 26.2%, consisting of 23.2% for patients matched by DNA analysis and 31.6% for matching by RNA expression. The cohort had a median PFS of 2.1 months (1.94 for those matched by DNA analysis and 2.43 for patients matched by RNA expression). The investigators found that 23 of the 107 patients (or 22.4%) had PFS2 that exceeded PFS1 by more than 50% — thus falling short of the primary endpoint.
Nevertheless, the investigators found the results encouraging.
“The strategy employed in WINTHER resulted in a higher proportion of patients treated than in many precision medicine trials,” the American Society of Clinical Oncology chief medical officer and study co-author Richard Schilsky, MD, said in a statement. “Previous studies have identified potential treatments for 5% to 25% of patients based on DNA profiling alone. Our findings represent an important step toward delivering on the true promise of precision medicine in oncology.”
Looking ahead to future research and goals, Kurzrock told MedPage Today that combining DNA and RNA analysis into a single test is “absolutely feasible and desirable.” Developing a blood test that performs both types of analyses might be possible but requires considerable more work to get a definitive answer, she added.
The study was supported by the European Union, the ARC Foundation of France, Pfizer Oncology, Lilly France SAS, and Novartis.
First author Jordi Rodon and multiple coauthors disclosed relationships with the pharmaceutical industry and other commercial interests.