Phase II Trial Shows Feasibility of Customized Adjuvant Treatment in NSCLC, but Phase III Trial Canceled Due to Unreliability of ERCC1 Readouts
In the phase II Tailored Postsurgical Therapy in Early-Stage NSCLC (TASTE) trial (IFCT-0801), reported in the Journal of Clinical Oncology, Wislez et al examined the feasibility of customized adjuvant treatment based on EGFR mutation status and expression of ERCC1 (excision repair cross-complementation group 1), a predictor of cisplatin response, in patients with non–small cell lung cancer (NSCLC). Although the trial met its primary endpoint of ≥ 80% of patients being able to start adjuvant chemotherapy within 2 months of surgery, a phase III trial of the customized approach was cancelled due to unreliability of ERCC1 immunohistochemical readouts.
The relationship between ERCC1 expression and cisplatin resistance was confirmed by ERCC1 immunohistochemical analysis in NSCLC patients in the IALT (International Adjuvant Lung Cancer) trial, in which adjuvant therapy significantly prolonged survival of patients with ERCC1-negative tumors vs ERCC1-positive tumors compared with observation. Among patients not receiving adjuvant chemotherapy, survival was longer in those with ERCC1-positive vs ERCC1-negative tumors, suggesting that ERCC1-positive patients can be spared chemotherapy.
Study Details
In the study, 150 patients with completely resected nonsquamous cell stage II or IIIA non-N2 tumors were randomly assigned to four standard-dose courses of cisplatin plus pemetrexed (Alimta) (n = 74) or customized treatment (n = 76) consisting of erlotinib (Tarceva) at 150 mg/d for 1 year in patients with activated EGFR mutations, four courses of cisplatin/pemetrexed in ERCC1-negative patients, and follow-up in ERCC1-positive patients. The primary objective was to start adjuvant therapy by day 61 after surgery, with biomarker information being available no later than day 56.
Success Rate
All patients in the standard treatment group received cisplatin/pemetrexed. In the customized treatment group, 53 (70%) received cisplatin/pemetrexed, 7 (9%) received erlotinib, and 16 (21%) underwent follow-up. Among all patients, chemotherapy was initiated before day 61 with biomarker results available in 120 patients, yielding an overall success rate of 80%, including rates of 77% in the standard treatment group and 83% in the customized treatment group.
Discrepant ERCC1 Immunohistochemical Findings
However, the investigators observed that their ERCC1 immunohistochemical results significantly differed from those in the IALT analysis, with 25.3% of patients being ERCC1-positive in the current study compared with 44% in the IALT study. Restaining of the IALT samples with the 8F1 antibody commercially available in 2011 showed discrepancies between the ERCC1 [immunohistochemical] results obtained in IALT in 2006 and those obtained in the current study.
The investigators stated, “The current batches of the monoclonal antibody 8F1, as well as all other commercial antibodies, are unable to make the distinction between the four isoforms of the ERCC1 protein, whereas only one isoform has full capacities for nucleotide excision repair. The ERCC1 [immunohistochemical] readouts were thus found to be unexpectedly unreliable.” The phase III portion of the TASTE program was thus cancelled.
The investigators concluded, “The primary end point of the trial was met, demonstrating the feasibility of a national biology-driven trial in the adjuvant NSCLC setting. Nevertheless, the phase III part was canceled because of the unreliability of the ERCC1 immunohistochemical readouts.”
Marie Wislez, MD, PhD, of Assistance Publique des Hôpitaux de Paris, is the corresponding author for the Journal of Clinical Oncology article.
The study was supported by the French Cooperative Thoracic Intergroup, Institut National du Cancer bioTASTE grants, and grants from Eli Lilly and Roche. For full disclosures of the study authors, visit jco.ascopubs.org.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
