'Genome-Driven' Cancer Drugs Treat Small Minority of Patients

Pam Harrison

May 03, 2018

The use of genome-driven cancer therapies that target aberrations on tumor cells is limited to a small percentage of patients with advanced or metastatic cancer, and those numbers call into question the value of further research into targeted therapies, according to the authors of a retrospective, cross-sectional study.

"The enthusiasm for these therapies is very very high and a lot of this enthusiasm is driven by the fact that these therapies make a lot of sense — they should work and some do work, although others are not as great," senior author, Vinay Prasad, MD, MPH, Oregon Health & Science University, Portland, told Medscape Medical News.

"The purpose of our paper was to ask how many people actually benefit from these therapies, and we found that about 9 out of 100 people who present with cancer could get one of these drugs and maybe 5 out that 100 would actually benefit from treatment by having their tumor shrink, which is actually a generous definition of benefit," Prasad observed.

The study was published online April 17 in JAMA Oncology.

Investigators evaluated 31 drugs that had been approved by the US Food and Drug Administration for 38 indications between 2006 and 2018. The data set included trastuzumab, imatinib, gefitinib, erlotinib, and cetuximab, all of which were approved before 2006.

The drugs were "genome-targeted" or "genome-informed."

Genome-targeted drugs were defined as drugs used on the basis of findings from a genomic test where the drug targets the aberration identified on testing.

Genome-informed drugs were defined as all genome-targeted drugs as well as any drug given after a patient has undergone genomic testing, regardless of whether the drug targets abnormalities identified on the test. 

Almost three quarters (73.7%) of the drugs evaluated by the research team were genome-targeted drugs; the remaining 10 agents were classified as genome-informed.

Patients were judged to be eligible for a given drug if they exhibited both the type of tumor and the genetic abnormality for which a drug is indicated.

"We collected annual mortality statistics by cancer type from the American Cancer Society to ascertain the number of patients who died annually and could have benefited from genome-targeted or genome-informed therapies," the researchers observe.

In the study, the investigators estimated that the number of patients eligible for genome-targeted therapy in 2006 was 28,729 out of a total 564,830 patients with metastatic cancer, or 5.09%.

By 2018, the number of patients increased to 50,811 of 609,640, or 8.33%.

However, in 2006, only an estimated 0.70% of US patients with cancer responded to genome-targeted therapy and therefore benefited. In 2018 that number increased to 4.90%.  

Using the category of genome-informed therapy, the researchers found that slightly more patients benefited from treatment, at 1.3% in 2006 and 6.6% in 2018.

The researchers also analyzed their dataset assuming that "real-world" response rates would be 10% or 20% lower than response rates reported in clinical trials.

On the basis of these assumptions, they estimated that approximately 4% of "real-world" patients would benefit from genome-targeted therapies in 2018 while between 5% and 6% of patients would benefit from genome-informed therapies in the same year.

"Right now, we are investing heavily in immunotherapy and heavily in genomic therapy, but in other categories of drugs, such as cytotoxic drugs, we have stopping investigating in them," Prasad cautioned.

"But it's foolish to do this — we need to have the vision to look beyond the fads we live by in cancer medicine and do things in a broader way," he said.

"So I support broader funding because you have to sustain efforts even when things are not in vogue if you want to make progress," Prasad concluded.

Unbridled Enthusiasm

Asked by Medscape Medical News to comment on the study, Ian Tannock, MD, PhD, Princess Margaret Cancer Centre and the University of Toronto in Toronto, Ontario, Canada, agreed with Prasad that "unbridled enthusiasm" for genome-driven oncology is "somewhat misplaced."

"The problem is that while there are some successes, most of these are relatively simple," Tannock elaborated.

For example, drugs such as trastuzumab, which targets HER2 in breast tumors, or vemurafenib (Zelboraf, Genentech) and dabrafenib (Tafinlar, Novartis), which target BRAF mutations in melanoma, do represent a "substantial advance" in the treatment of these malignancies.

Nevertheless, Prasad and colleagues considered both of these treatments in their analysis, and even so, "they represent a small percentage of the total benefit," Tannock said.

A greater problem in Tannock's view is that several companies are already marketing next-generation sequencing to convince patients with cancer to have their tumors sequenced.

"These patients are spending somewhere in the range of $5000 to get their tumors sequenced in the belief that this will make their tumors more treatable, which is absolutely wrong," Tannock emphasized.

"This may be appropriate in a research setting, but current trials that have looked at this have shown very, very minimal benefit from this because sequencing rarely changes anything," he cautioned.

Commenting further on the study, H. Jack West, MD, medical director, Thoracic Oncology Program, Swedish Cancer Institute, Seattle, Washington, told Medscape Medical News that he believes Prasad and colleagues are doing the right thing by challenging the medical community to step back and examine the evidence supporting genome-driven oncology and ask whether they are meeting the "sensationalism" of it.

"He is not saying that nobody benefits from it, what he is saying is: from all you hear about precision medicine, the majority of patients with cancer think they have some decodable genetic mutation with a great treatment," West elaborated.

The reality is that patients eligible for genomic therapies are actually in the single or low double digits, as Prasad and colleagues showed, and does not merit such outsized expectations.

"It's not to say that these targeted therapies don't work — the data the study alludes to suggests that for a minority of patients, they can work very well," West reaffirmed.

"But we should not be using hyperbole by implying that the majority of patients have some incredibly treatable mutation because the progress we have made in this arena has been steady, methodical, and slow," he added.

"And even though the talk is all about immunotherapy and targeted therapies now, we have to recognize that less glamorous models of chemotherapy actually still do a great job at improving outcomes for patients with a lot less fanfare."

Prasad is a coauthor of Ending Medical Reversal: Improving Outcomes, Saving Lives, for which he receives royalties. Tannock has disclosed no relevant financial relationships. West reports that he has served as a director, officer, partner, employee, advisor, consultant, or trustee for Genentech/Roche, Merck & Co  Inc, and Clovis Oncology.

JAMA Oncol. Published online April 17, 2018. Abstract

For more from Medscape Oncology, follow us on Twitter:  @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: