Guiding treatment for non–small-cell lung cancer (NSCLC) using expression levels of the proteins RRM1 and ERCC1 led to no difference in either progression-free or overall survival, according to a study presented at the American Society of Clinical Oncology (ASCO) annual meeting.
CHICAGO-Guiding treatment for non–small-cell lung cancer (NSCLC) using expression levels of the proteins RRM1 and ERCC1 led to no difference in either progression-free or overall survival, according to a study presented at the American Society of Clinical Oncology (ASCO) annual meeting.
Data from laboratory and early-stage studies has shown that “there is an association between levels of expression of ERCC1 and the response rate of platinum agents,” said Gerold Bepler, MD, PhD, of the Karmanos Cancer Institute at Wayne State University in Detroit, who presented the study’s results. Furthermore, “as RRM1 levels increase, resistance to gemcitabine increases, and the change in resistance is approximately 10-fold. The higher the levels of RRM1, the less effective gemcitabine is at reducing tumor burden.”
The investigators conducted a trial in which treatments were thus guided by levels of those proteins. There were 183 patients in an experimental arm and 92 patients in a control arm; the control group all received gemcitabine and carboplatin. The experimental arm was divided into four subgroups based on protein expression: Low levels of both proteins received the same regimen as the control group; low RRM1 and high ERCC1 received gemcitabine and docetaxel; high RRM1 and low ERCC1 received docetaxel and carboplatin; and patients with high levels of both received docetaxel and vinorelbine. The distribution of patients to those groups did not differ from expected rates (P = .2), and there was a significant correlation between expression levels of the two proteins (R = 0.394; P < .001), which Dr. Bepler said was expected.
There was no difference seen with regard to the primary endpoint of progression-free survival, which was 6.1 months in the experimental arm and 6.5 months in the control group (P = .181). Overall survival was also similar, at 11 months and 11.3 months, respectively. Interestingly, the only significant difference in the trial was seen in progression-free survival for the experimental group with low levels of both proteins compared with those in the control group with low levels of both (5 months vs 8.1 months). The investigators looked for sources of bias to explain this, as those two groups of patients had the same protein expression levels and received the exact same treatment combination.
No bias was found, but the session’s discussant, Lecia V. Sequist, MD, MPH, of Massachusetts General Hospital in Boston, said there were some potential biases, such as differing numbers of men and women, as well as rates of adenocarcinoma that could explain it. She also noted that the subgroup analyses were not preplanned.
“The authors should be congratulated on completing a logistically complex trial,” she said, but agreed with Dr. Bepler’s conclusion that “there is really no alternative except to interpret the study as a negative study.”
Dr. Bepler did say that this shows at least that treatment assignment based on molecular analysis of tumor specimens is feasible. “Most patients accept a rebiopsy of the tumor for molecular studies,” he said, noting a rate of 89% in those patients that did require a second biopsy.