SAN FRANCISCO – Pretreatment expression of ERCC1 in localized esophageal and gastroesophageal adenocarcinomas is a marker for outcomes among patients given trimodality therapy that includes oxaliplatin-based chemoradiation, investigators reported from a prospective study.
Fully 58% of the 55 patients studied from the Southwest Oncology Group (SWOG) S0356 trial had tumors expressing a high level of mRNA for ERCC1, a key gene in the repair of platinum- and radiation-induced DNA damage.
Compared with their counterparts whose tumors had low expression, these patients were nearly three times more likely to experience progression and more than twice as likely to die, according to results reported at a meeting on gastrointestinal cancers sponsored by the American Society of Clinical Oncology.
Dr. Pierre O. Bohanes
"ERCC1 mRNA level is a very promising pretreatment biomarker in patients with localized esophageal and gastroesophageal adenocarcinoma treated with trimodality treatment," asserted lead investigator Dr. Pierre O. Bohanes. "Based on these and published data, the SWOG is planning a prospective biomarker-driven clinical trial."
"In contrast to the growing number of predictive biomarkers for anticancer agents, there are no established biomarkers to select patients who will benefit most from chemoradiation," he noted. "Utilization of predictive biomarkers to select therapy should lead to higher cure rates."
The phase II trial, which was largely community based, enrolled patients with clinical stage II or III esophageal or gastroesophageal junction adenocarcinoma. They underwent tumor biopsy, then chemoradiation (oxaliplatin, 5-fluorouracil, and 45-Gy external beam radiation), then surgery, and finally more chemotherapy (oxaliplatin and 5-fluorouracil).
Response Genetics Inc., manufacturer of several biomarker assays, analyzed tumor ERCC1 mRNA expression for the 55 patients who had sufficient pretreatment tumor tissue. Laser-capture microdissection was used to ensure that only tumor cells were analyzed.
"ERCC1 has been shown to be a critical gene in DNA repair," Dr. Bohanes explained. "ERCC1 is involved in the nucleotide excision repair pathway, which recognizes and removes platinum-induced DNA adducts." Higher expression is associated with resistance to platinum compounds, including cisplatin, oxaliplatin, and carboplatin.
"Also more recently, ERCC1 has been shown to be involved in the double-strand break repair pathway, which repairs radiation-induced damage," he further observed.
Tumors were classified as having a high or low level of ERCC1 expression using a cutoff mRNA level of greater than 1.7 for high expression, as established in previous studies.
The 55 patients studied had a median age of 64 years. Most were men (93%) and white (94%), and had esophageal tumors (62%). One-fourth of them were found to have a pathological complete response at the time of surgery.
The median duration of follow-up was 36.8 months, reported Dr. Bohanes, a research fellow in medical oncology at the University of Southern California in Los Angeles.
Results showed that patients having high vs. low tumor expression of ERCC1 had a nearly tripled risk of progression (hazard ratio, 2.97; P = .006). The median duration of progression-free survival was 14.8 months for the former but was not reached for the latter. Corresponding 2-year rates of progression-free survival were 17% and 67%.
Similarly, patients with tumors having high vs. low expression of ERCC1 had a more than doubling of the risk of death (HR, 2.32; P = .047).The median duration of overall survival was 22.4 months for the former but was not reached for the latter. Corresponding 2-year rates of overall survival were 37% and 72%.
Expression of ERCC1 was not associated with pathological complete response. Also, expression of a host of other genes – XPD and RRM1 (associated with DNA repair), GSTP1 (associated with platinum detoxification), and TS, TP, and DPD (associated with 5-fluorouracil metabolism) – was not associated with any of the outcomes studied.
"Biomarker studies are feasible within cooperative groups," commented Dr. Bohanes, but adequate tissue was available for only 55 of 92 eligible patients. Hence, "future trials need to request additional endoscopic biopsies to allow for sufficient tumor tissue collection."
A study shortcoming was an inability to determine whether ERCC1 expression correlates with disease stage, he acknowledged. "The design of this study unfortunately did not require preoperative endoscopic ultrasound because it was thought that it would have hampered the recruitment of this trial," he said. "And often, in the community setting, endoscopic ultrasound is not available."
Also, the study could not determine whether ERCC1 expression is a prognostic marker. "We would need a control arm without platinum agent," he noted.
Dr. Heinz-Josef Lenz
"I think that the patients with high expression are not benefiting from this treatment" and should be treated with alternative treatments, commented Dr. Heinz-Josef Lenz, senior investigator and chair of the gastrointestinal oncology program at the University of Southern California. "We have choices, and I think that would certainly be part of the new concept and design going forward."