Individuals with oral premalignant lesions (OPLs) are at heightened risk of developing head and neck squamous cell carcinoma (HNSCC) or oral cancer. Because epidermal growth factor receptor (EGFR) plays a critical role in oral epithelial carcinogenesis, there has been hope that EGFR inhibitors could have a role in chemoprevention.
That hope has been dashed somewhat by negative results from a clinical trial with the EGFR inhibitor erlotinib (Tarceva, Osi Pharmaceuticals, Inc), published online November 5 in JAMA Oncology.
"Although we are disappointed that erlotinib was not effective, this is the first precision medicine trial in cancer prevention and will change the way we do prevention trials," Scott M. Lippman, MD, of the Moores Cancer Center of the University of California, San Diego (UCSD), told Medscape Medical News.
"Despite the promise of precision chemoprevention, erlotinib now joins the list of agents that have been tested and failed to prevent oral carcinogenesis," state Julie E. Bauman, MD, MPH, of the University of Pittsburgh Cancer Center, in Pennsylvania, and Jennifer Grandis, MD, of the University of California, San Francisco, in an accompanying editorial.
"However, by prospectively validating a molecularly selected, high-risk population and demonstrating the feasibility of a cancer-free survival endpoint, the EPOC [Erlotinib Prevention of Oral Cancer] study has initiated a new epoch in HNSCC cancer chemoprevention trials," write the editorialists.
Why EPOC Failed
"EPOC was the first chemoprevention study that studied whether erlotinib would prevent oral cancer in patients at high risk for oral cancer," Dr Lippman told Medscape Medical News.
The team selected patients with well-defined premalignant lesions, performed a biopsy to molecularly profile the lesions, and then chose patients with loss of heterozygosity (LOH) to participate in the trial.
High-risk LOH profile was defined as LOH at 3p14 and/or 9p21 in individuals with a history of cancer or LOH at 3p14 and/or 9p21 and an additional chromosome site (17p, 8p 11p, 4q, or 13q) in individuals without a history of oral cancer. These LOH loci are associated with several tumor suppressor genes.
The trial randomly assigned patients with LOH to receive erlotinib or placebo in a double-blinded fashion, Dr Lippman explained to Medscape Medical News.
However, the incidence of cancer-free survival — the primary endpoint of the study ― was not significantly different between the two arms of the study, and thus EPOC found that erlotinib is not effective at chemoprevention.
"The EPOC study was based on convergent clinical and preclinical evidence," the editorialists point out. Another drug targeting EGFR, the monoclonal antibody cetuximab (Erbitux, ImClone Systems Incorporated), is clinically useful in the treatment of HNSCC and has been shown to improve survival. And erlotinib prevented oral dysplasia and invasive carcinoma in a murine model of oral cancer, they explained.
"Given this compelling evidence, why did the EPOC study have negative results?" the editorialists ask.
"EPOC likely failed because of a disconnect between LOH as a prognostic biomarker and lack of preclinical evidence that LOH serves as a predictive biomarker for erlotinib treatment," Dr Grandis told Medscape Medical News.
Although EGFR gene copy number was a good prognostic marker in oral cancer in EPOC, it was not a biomarker predictive of treatment with EGFR inhibitors, Dr Grandis explained.
Although the EPOC investigators provide a plausible mechanistic link to the specific LOH loci, which deleted tumor suppressor genes, included in the EPOC eligibility, the "the specificity of erlotinib for these LOH profiles was not mechanistically defined in preclinical models and perhaps was the primary reason that promising preclinical results failed to translate to the clinic," Dr Bauman and Dr Grandis write in their editorial.
Although EPOC also found a positive correlation between the occurrence of rash — a toxicity associated with erlotinib — and improved cancer-free survival, the high rate of dose reductions due to toxic effects (45% for individuals receiving erlotinib vs 1 % for patients receiving placebo) may not be tolerated in this setting, the EPOC investigators explain.
Dr Grandis agreed. "Erlotininb has toxic side effects, which may be acceptable for patients with cancer. For patients who do not have cancer, any toxicity is harder to accept; it is easy to see why discontinuation or dose reductions were required in close to half the patients," she told Medscape Medical News.
"By definition, chemoprevention occurs in healthy people at risk for, but without evidence of, cancer. Thus, the toxicity profile of candidate agents is often prohibitive, particularly when chronic exposure is required for prevention by delay," Dr Bauman and Dr Grandis write.
EPOC: The Study and Results
Patients with LOH were enrolled with or without a history of oral cancer, which was the basis for stratifying patients in the two study arms.
Of 375 individuals with classifiable LOH profiles, 121 were LOH negative and 254 were LOH positive; of individuals with an LOH-positive profile, 104 did not undergo randomization; 150 were randomly assigned to receive erlotinib (n = 75) or placebo (n = 75). Patients without LOH were observed for oral cancer progression.
Individuals who were administered erlotinib received 150 mg/day orally (the full, standard therapeutic dose used for lung cancer, the approved indication for the drug); dose reductions to 100 mg/day orally and then to 50 mg/day orally were permitted to address protocol-defined toxicity.
Patients were treated for 1 year and were followed for a median of 35 months.
EPOC is the first study of oral premalignant lesions that used cancer as an endpoint, the researchers point out.
The 3-year cancer-free survival was 70% for individuals receiving erlotinib and 74% for those receiving placebo (hazard ratio: 1.27; 95% confidence interval: 0.68 - 2.38; P = .45).
Analysis of subgroups showed no significance with erlotinib for patients with mild/moderate dysplasia, severe dysplasia, or increased EGFR gene copy number.
EPOC provided some additional insights. First, 3-year cancer-free survival was significant for patients without LOH (87% vs 74% for the LOH-positive group; P < .01), establishing prospectively that LOH is as a prognostic marker. In addition, LOH positivity together with an increased EGFR gene copy number identified individuals with the highest risk for oral cancer.
"One of the greatest challenges in developing chemopreventive agents is to identify the population at highest cancer risk," William N. William, MD, of the MD Anderson Cancer Center, who led the enrollment and design of EPOC, said in an press release issued jointly by the MD Anderson Cancer Center and UCSD.
"Not all patients with an oral premalignant lesion will develop oral cancer. By using a molecular test that can identify those at highest risk, we can focus preventive efforts on these specific individuals," he added.
"The EPOC trial demonstrated prospectively the role of LOH as a marker of cancer risk in OPLs and supports its use as a prognostic tool in clinical practice," Dr Williams told Medscape Medical News.
"The novel approach in the EPOC trial of selecting a molecularly defined high-risk population most likely to benefit, allowing the first streamlined OPL trial designed with the clinically meaningful definitive endpoint of invasive cancer, is a new paradigm for future precision cancer chemoprevention trials," Dr William told Medscape Medical News.
The editorialists identify three dominant barriers to the success of chemoprevention trials: defining a high-risk population in whom HNSCC event rates necessitate an intervention; long-term administration of a toxic agent to healthy individuals without cancer; and successful translation of preclinical models of oral cancer to the clinic.
Although EPOC was successful in addressing the first barrier, its failure may be attributed to its inability to address the last two barriers, the editorialists opine.
The work was funded by OSI Pharmaceuticals, which also provided drug supply. Dr Lippman, Dr William, and Dr Grandis report no relevant financial relationships. Two study investigators are on advisory boards of several pharmaceutical companies, which are listed in the original article.
JAMA. Published online November 5, 2015.
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου