Not Yet Ready for Prime Time: Lessons Learned From a Failed Molecular Marker-Based Lung Cancer Trial

OncologyLive, August 2014, Volume 15, Issue 8

Scarcely a week goes by without a report in the peer-reviewed medical literature suggesting that a novel somatic genomic alteration or a specific normal polymorphism is potentially relevant to achieving a desirable clinical outcome within cancer medicine.

Maurie Markman, MD

Editor-in-Chief of OncologyLive

Senior vice president for Clinical Affairs and National Director for Medical Oncology Cancer Treatment Centers of America, Eastern Regional Medical Center

Scarcely a week goes by without a report in the peer-reviewed medical literature suggesting that a novel somatic genomic alteration or a specific normal polymorphism is potentially relevant to achieving a desirable clinical outcome within cancer medicine. These outcomes include both enhanced efficacy, such as improved time to disease progression or overall survival, and reduced toxicity associated with a particular antineoplastic strategy.

Inherent in these provocative, even paradigm-changing reports is the development of a well-characterized biological marker that reliably and reproducibly defines (predicts) the patient population where a particular strategy has a high probability of resulting in clinical benefit (eg, tyrosine kinase inhibitors of EGFR in the presence of a documented EGFR mutation in patients with non—small cell lung cancer [NSCLC]) or the absence of clinical utility (eg, anti-EGFR antibody therapy for metastatic colon cancer in the presence of a KRAS mutation).

Considering the potential impact on clinical outcomes, it is perhaps not surprising that there often appears to be a greater focus on the therapeutic implications of a particular molecularly defined event than there is—or should be—either on the underlying clinical validity of the molecular target itself, or on the specific diagnostic tool being employed to define the target within a patient population and then document the presence of that target in a particular individual.

Table. Two Findings of ERCC1 Expression in NSCLC


Patients, N

ERCC1+ % (n)

ERCC1— % (n)



25.3% (38)

60% (90)



44% (335)

56% (426)

aERCC1 status could not be determined in 14.7% of patients.

ERCC1 indicates excision repair cross-complementation group 1;

NSCLC, non—small cell lung cancer.

It is important to acknowledge that there are serious risks associated with underestimating the difficulty of finding and objectively validating the clinical relevance of a proposed biomarker, including the fortunately uncommon but real-world concern for scientific and financial misconduct in such a complex arena.1

Unfortunately, the problems that arise in this field of research and the failure of specific investigative efforts to confirm the utility of such diagnostic strategies are less frequently reported in the peer-reviewed literature than one might reasonably expect. Therefore, it is refreshing to see a research team overcome the obvious disappointment when an endeavor fails to validate a proposed technique and report the results in an honest analysis with suggested strategies for moving the given hypothesis forward.

French Trial Tackles Testing Nuances

A recent report of a trial examining a “customized” adjuvant chemotherapy strategy for patients with stages II or IIIA NSCLC from investigators in France clearly falls into this category of greater transparency.2

The hypothesis in this randomized phase II study, the Tailored Postsurgical Therapy in Early- Stage NSCLC (TASTE) trial, was that delivering adjuvant antineoplastic treatment based on a laboratory- defined molecular profile would improve outcomes compared with administering standard- of-care chemotherapy to all patients. Specifically, researchers sought to evaluate the utility of considering the expression levels of excision repair cross-complementation group 1 (ERCC1), a DNA-repair biomarker, in predicting response to platinum-based therapies.

All patients randomized to the control arm received adjuvant cisplatin/pemetrexed as standard- of-care therapy. In the experimental arm, patients whose cancers were found to contain an EGFR mutation received erlotinib. In the absence of a documented EGFR mutation, patients whose tumors were found to be ERCC1 “negative,” suggesting responsiveness to platinum therapy, received adjuvant cisplatin/pemetrexed, while those whose tumors were ERCC1 “positive” went on to observation.

The investigators found it was possible to successfully conduct the trial; that is, treatment regimens were tolerable and researchers were able to obtain biomarker testing in a timely manner. Eighty percent of the patients entered into the study initiated adjuvant therapy within 2 months of surgical tumor resection.

However, the investigators elected not to proceed to a definitive phase III study due to the unreliability of the ERCC1 testing procedure. The antibody used in the TASTE trial produced a substantially lower percentage of tumors that were ERCC1 “positive” compared with a previously reported study, the International Adjuvant Lung Cancer Trial Biology (IALT Bio) study.3 (Table). Of note, problems with commercial antibodies suggested to predict for nucleotide excision repair have recently been identified by others.4

Undoubtedly, defining the inherent chemosensitivity of an individual cancer prior to the administration of adjuvant therapy, either systemic cytotoxic drugs or more targeted antineoplastics, remains an important therapeutic concept. However, it is critical that any biomarker employed within the clinical domain to define such sensitivity be able to produce both highly reproducible and objectively valid results.

Maurie Markman, MD, editor-in-chief, is president of the Medicine & Science at Cancer Treatment Centers of America and clinical professor of Medicine, Drexel University College of Medicine.


  1. Goldberg P. Duke halts third trial; coauthor disputes claim that data validation was blinded. The Cancer Letter. 2009;35(39):1.
  2. Wislez M, Barlesi F, Besse B, et al. Customized adjuvant phase II trial in patients with non—small-cell lung cancer: IFCT-0801 TASTE. J Clin Oncol. 2014;32(12):1256-1261.
  3. Olaussen KA, Dunant A, Fouret P, et al. DNA repair by ERCC1 in non—small cell lung cancer and cisplatin-based adjuvant chemotherapy. N Engl J Med. 2006;355(10):983-991.
  4. Friboulet L, Olaussen KA, Pignon J-P, et al. ERCC1 isoform expression and DNA repair in non—small-cell lung cancer. N Engl Med. 2013;368(12):1101-1110.