Register TodayEarn Free CME Credits by reading the latest medical newsin your specialty.Sign Up
A DNA repair biomarker thought to predict benefit from platinum-based chemotherapy in non-small cell lung cancer (NSCLC) doesn't actually do that good a job, a validation study showed.
The ERCC1 protein expression level didn't predict a boost in overall survival (OS) from adjuvant cisplatin (Platinol)-based chemotherapy compared with observation alone in two clinical trials (P=0.23 for interaction), Jean-Charles Soria, MD, PhD, of Institut Gustave-Roussy in Villejuif, France, and colleagues found.
In fact, any hint of utility in distinguishing chemotherapy response actually came from the ERCC1-positive group, rather than the ERCC1-negative group as suggested in the initial trial proposing the protein as a predictive biomarker.
The problem appeared to be both technical and due to the inability of the assay to distinguish the key form of the protein for DNA repair, the group reported in the March 21 issue of the New England Journal of Medicine.
"Currently available antibodies do not have adequate discrimination for therapeutic decision making regarding cisplatin-containing treatment in patients with NSCLC, which requires the specific detection of the unique functional isoform of ERCC1 -- ERCC1-202," they wrote.
The other three isoforms of ERCC1 (excision repair cross-complementation group 1) protein aren't as critical in fighting the cytotoxic effect of platinum chemotherapy.
However, the isoforms are so similar in shape that manufacturing a specific antibody that could pick out only ERCC1-202 would be a daunting challenge, Grace Dy, MD, of Roswell Park Cancer Institute in Buffalo, N.Y., explained in an interview.
"That does not mean that ERCC1 itself is totally useless," she told MedPage Today. But "it is a reminder to us that getting a biomarker validated and tested is a tortuous and difficult path."
The researchers agreed that another route may have to be found to make use of ERCC1 as a biomarker for platinum chemotherapy response.
Most trials looking at ERCC1 expression in NSCLC, since the initial demonstration of its predictive value in a subanalysis of the International Adjuvant Lung Cancer Trial (IALT), has used immunohistochemistry with a mouse monoclonal antibody.
But no consensus has been reached for the best was to evaluate ERCC1 expression, Soria's group noted.
"Functional assays may be required to detect an authentic prognostic influence of DNA repair proteins," they concluded.
They dug into the specifics of the immunohistochemistry method by analyzing a validation set of samples from 494 patients in two independent phase III trials -- the National Cancer Institute of Canada Clinical Trials Group JBR.10 and the Cancer and Leukemia Group B 9633 trial from the Lung Adjuvant Cisplatin Evaluation Biology project.
Those trials did not confirm any OS advantage from adjuvant platin-based chemotherapy versus no chemotherapy controls among tumors that stained negative for ERCC1, with a mortality hazard ratio of 1.16 that wasn't even close to statistical significance (P=0.62).
Also contrary to what would have been expected from the IALT results, there was some hint of chemotherapy benefit within the ERCC1-positive group. The HR of death from any cause was 0.78 versus controls, although still not a statistically significant trend (P=0.09).
To see what might be behind these differences, the researchers re-analyzed the original IALT Biology subanalysis.
Repeat staining of the entire set of tumor-block samples indicated a big shift in ERCC1 status, with 78% scoring positive compared with the reported 44%.
That changed the results too, with ERCC1 no longer a predictive biomarker for OS benefit from chemotherapy (P=0.53 for interaction).
The validation study used a different batch of the mouse antibody than used in 2006 in the original study.
But different cutoffs for positivity didn't improve prediction, the researchers noted.
Testing 14 other commercially-available monoclonal antibodies for ERCC1 detection indicated that none were specific for only one isoform; all responded the same to at least three of the four isoforms.
Quantitative real time-PCR analysis of mRNA levels of ERCC1 could distinguish response to two isoforms -- the key 202 isoform plus 204 -- but didn't prove to be any more predictive of platin chemotherapy response.
The researchers suggested this as one explanation for their results because "such cross-reactivity could introduce unexpected variability in staining in validation studies" and "lead to potential artifacts, with discrepant results."
The study was supported by an unrestricted grant from Eli Lilly and by grants from Institut National du Cancer Programme National d'Excellence Spécialisé and Programme Hospitalier de Recherche Clinique; by La Ligue Nationale contre le Cancer and an unrestricted grant from sanofi-aventis; by a contract from the European Union Seventh Framework Program under a grant agreement; and by a translational research fellowship from Roche.
Soria reported grants to his institution from Eli Lilly and sanofi-aventis and being named on a patent regarding ERCC1 expression in predicting response for cancer chemotherapy, although without any revenues to him or his institution.
Crystal Phend
Staff Writer
Crystal Phend joined MedPage Today in 2006 after roaming conference halls for publications including The Medical Post, Oncology Times, Doctor's Guide, and the journal IDrugs. When not covering medical meetings, she writes from Silicon Valley, just south of the San Francisco fog.
No comments:
Post a Comment