One of the most interesting sessions I attended at this year’s American Society of Clinical Oncology (ASCO) annual meeting in Chicago was the Clinical Science Symposium (CSS) on the next generation of EGFR inhibitors.
We’ve previously written on the blog about the data for AZD9291 and CO-1686 presented at ASCO, but the CSS also featured an informative discussion by Larry Schwartz, MD, Professor of Radiology at Columbia University Medical Center which raised questions about how we should evaluate new lung cancer drugs.
In a presentation entitled, “Getting the Right Drug to the Right Patient Faster,” Schwartz who is a diagnostic radiologist, discussed and critiqued abstract 8012 by Dr Gideon Michael Blumenthal and colleagues at the U.S. Food and Drug Administration (FDA).
A meta-analysis of fifteen trials involving 12,534 patients (median N = 698) from nine experimental agents (tyrosine kinase inhibitor = 5, chemotherapy = 2, monoclonal antibody = 2) submitted to the FDA for treatment of metastatic non-small cell lung cancer (NSCLC) cancer in initial or supplemental New Drug or Biologics License Applications since 2003 was performed by Blumenthal and colleagues.
Their analysis showed a strong correlation (R² of 0.89) between overall response rate (ORR) and progression-free-survival (PFS) but only a weak or no correlation between ORR and overall survival (OS) (R² of 0.07) or between PFS and OS (R² of 0.09).
Dr Blumenthal noted in his conclusion that further work is ongoing to corroborate these findings given the lack of correlation between OS and ORR could have been due to high cross-over, under-power and long post-progression survival.
He went on to note that what the findings do show is that “a drug with a large effect on ORR is likely to have a large effect on PFS, conversely a drug with a small ORR may have a small effect on PFS.”
The debate around objective response and outcomes is a very interesting one, as is the drive to find better biomarkers of response to improve chances of clinical trial success.
Subscribers can login to read more about our insights on this important topic.