September 24, 2018 by Dale Yuzuki
Being unaware of oncology precision medicine or being oversold on oncology precision medicine are both undesirable outcomes
There are two opposing forces at work in genetic and genomic testing for cancer: the first is making patients aware of new treatment options and companion diagnostics approaches in choosing these new treatments, and the second is oversimplifying the message in the sales and marketing of such options. In a recent JAMA Oncology editorial, Dr. Jack West of the Swedish Thoracic Hospital (Seattle WA) asks the question: “Can we integrate genomic testing without selling it?”
Broad-based genomic sequencing in advanced NSCLC
This editorial is based upon research reported in the same journal, looking at the association of broad-based genomic sequencing (which the paper shortens to ‘BGS’) in advanced NSCLC to survival benefit in the community oncology setting. This is an important topic due to the high number of NSCLC cases – an estimated 234,000 in the US for 2018 and 154,000 deaths; the 5-year survival rate for Stage IIIB is only 26%, and for Stage IV is 1%.
The retrospective study looked at data from a single (out of 191) US oncology practice in the community hospital setting that used the Flatiron Health Database, and patients had been diagnosed with either Stage IIIB/IV or unresectable nonsquamous NSCLC with at least one line of anticancer treatment. This study looked at 12-month mortality and overall survival (OS) from the start of first-line treatment examining 5,688 individual cases.
Not encouraging results for precision oncology
The results from this study were not heartening. Of the 875 had advanced NSCLC who received BGS, only 4.5% received targeted treatment, and 9.8% received routine EGFR/ALK targeted treatment. At 12 months, the overall survival (OS) rate of the patients receiving BGS was 49.2% and for the patients who received routine (EGFR and/or ALK) testing the survival rate was 35.9%. Statistical analysis concludes ‘broad-based genomic sequencing directly informed treatment in a minority of patients and was not independently associated with better survival’.
One confounding variable, the editorial points out, was the much higher rate of treatment by the BGS group to immune checkpoint inhibitors, which was then accommodated for in the subsequent instrument variable analysis. Dr. West points out ‘these findings should instill humility about the ability of BGS to transform cancer care’, and that ‘a largely overlooked potential harm of precision medicine is the probability of profound disappointment of clinicians and patients misled by a vague and unrealistic promise that BGS results will be relevant for a broad patient population’.
Another confounding effect is the effect of lack of available clinical trials in a community hospital setting, in addition to cost of treatment in those settings that limit availability of the targeted therapy to the appropriate patients.
An open debate
The popular press has taken note. In a recent editorial in the New York Times, health journalist Liz Szabo titles her piece “Are We Being Misled About Precision Medicine?”. The subhead reads, “Doctors and hospitals love to talk about the cancer patients they’ve saved, and reporters love to write about them. But deaths still vastly outnumber the rare successes.”
Earlier this year at the American Association for Cancer Research, a debate about precision oncology between Dr. David Hyman (Memorial Sloan Kettering NY) and Dr. Vinay Prasad (Oregon Health Sciences University OR) highlighted reasons for optimism (Dr. Hyman) as well as reasons for tempered enthusiasm (Dr. Prasad). “When you look at all of the data, it’s a sobering picture,” said Dr. Prasad. Dr. Hyman estimated only 15% of the 25,000 patient samples genomically tested at Sloan Kettering could be matched to an existing FDA-approved drug, and another 10-15% could be matched to an experimental drug undergoing tests in animals.
Dr. Prasad, looking at the same data with a more ‘glass half-empty’ view, he referred to his own work published during the AACR conference reaffirming the 15% that are eligible for FDA-approved therapy (his work found 15.1% of 614,000 US patients with metastatic cancer were eligible), but the data indicated only 6.6% likely benefited from such therapy.
The glass half-full is better than completely empty
Overall the measured improvement for advanced metastatic cancer due to genome-targeted therapy rose from 0.70% in 2006 to 4.90% in 2018, it is fair to call this response rate a ‘minority of patients’. This includes some 31 drugs with 38 FDA-approved indications for genome-targeted medicines. Let’s not ignore the fact however, that for those individuals (4.9% of 614,000 is over 30,000) it did make a difference in their cancer treatment.
The value proposition still exists for those 30,000 with advanced metastatic cancer; note that these patients have also likely to have been pre-treated. ‘Basket trials’ such as the NCI-MATCH aim for better randomized case-control studies to demonstrate the benefits of targeted therapy.
A balanced approach is in order
There is room for education of both physicians and patients, as well as for cautious optimism for precision oncology. The current 31 drugs for those 38 indications are sure to grow in the coming years, with an avalanche of targeted therapies in different phases of clinical trials.
Do you currently run a laboratory looking at genomic abnormalities in cancer? Are you interested in trying a targeted NGS enrichment approach that is more accurate, faster and less expensive? Contact us today.