CHICAGO – Patients with previously treated, advanced non–small cell lung cancer and high E-cadherin levels derived greater benefit when the investigational agent entinostat was added to erlotinib in a treatment strategy designed to overcome erlotinib resistance.
Dr. Robert Jotte
The data from a placebo-controlled, phase II trial are noteworthy because histone deacetylase inhibitors, such as entinostat, increase erlotinib sensitivity and prevent or delay resistance, which inevitably occurs in patients who are treated with erlotinib and other epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors.
In addition, patients with elevated E-cadherin levels account for 40% of the overall NSCLC population, making it an attractive clinical biomarker for directing therapy.
These preliminary data suggest that "there may be a subpopulation of patients for which entinostat may have the ability to overcome erlotinib resistance," Dr. Robert Jotte said at the Chicago Multidisciplinary Symposium in Thoracic Oncology, where he presented the findings in a late-breaking oral abstract.
Among 132 unselected patients in the phase II ENCORE 401 trial, outcomes were comparable for oral entinostat plus erlotinib (Tarceva) vs. erlotinib plus placebo, with a nonsignificant median progression-free survival difference of just 0.4 months and a median overall survival difference of 2.2 months, he reported.
Among 26 patients who were identified as having high E-cadherin levels, however, entinostat significantly increased overall survival, from 5.4 months with placebo to 9.4 months (hazard ratio, 0.36; P = .03), said Dr. Jotte, director of thoracic oncology at the Rocky Mountain Cancer Centers in Denver.
In contrast, 40 patients with low E-cadherin expression who received placebo survived a median of 7.0 months, compared with 4.4 months with entinostat (P = .55; HR, 1.25).
Median progression-free survival in the exploratory biomarker analysis also trended in favor of high E-cadherin patients who received entinostat vs. placebo (3.7 months vs. 1.9 months), but the difference did not reach statistical significance (HR, 0.55; P = .19).
Median progression-free survival among low E-cadherin patients was similar at 1.7 months with entinostat vs. 1.9 months with placebo (HR, 1.36; P = .36), said Dr. Jotte.
Invited discussant Dr. Pasi Jänne of the Dana-Farber Cancer Institute in Boston, said that preclinical data showing that erlotinib and entinostat prevent the emergence of resistance in EGFR-mutant cell lines (Cell 2010;141:69-80) provide a rationale for combining these agents. What isn’t clear is whether entinostat would restore erlotinib sensitivity in cell lines with a KRAS mutation or EML4-ALK translocation.
Dr. Jänne also observed that the current data contradict two other preclinical studies and a subset analysis from the TRIBUTE trial demonstrating that high E-cadherin expression is associated with erlotinib sensitivity. In ENCORE 401, however, patients with high E-cadherin expression who were treated with erlotinib and placebo had a shorter median overall survival than did their counterparts with low E-cadherin expression (5.4 months vs. 7.0 months).
Discrepancies between these data and the phase II data need to be resolved before additional prospective trials of erlotinib/entinostat are undertaken, Dr. Jänne said.
Syndax Pharmaceuticals, which is developing entinostat, plans to evaluate the drug in combination with erlotinib in further randomized trials to be initiated in 2011 in selected NSCLC patients with high levels of E-cadherin, according to a statement by Syndax president and CEO Dr. Joanna Horobin.
During a press conference at the meeting, Dr. Fred Hirsch, a professor of medicine with the University of Colorado Cancer Center in Aurora, also cautioned that an accepted standardized classification of E-cadherin expression needs to be established.
In ENCORE 401, immunohistochemistry staining values of +3 or greater were defined as high E-cadherin expression, and 0, +1, and +2 were defined as low E-cadherin expression.
In all, 132 patients who had progressed after one or two prior chemotherapies for stage IIIB/IV NSCLC were randomized 1:1 to erlotinib (150 mg once daily for 28 day) plus entinostat (10 mg on days 1 and 15 for 28 days), or to erlotinib and placebo. The majority was male (66%), had an ECOG (Eastern Cooperative Oncology Group) performance status of 0/1 (86%), had a history of smoking (85%) and had adenocarcinoma histology (43%). Only 11 of 78 patients who were tested had KRAS-mutant tumors.
Entinostat/erlotinib was tolerable with no unexpected adverse events and a manageable safety profile among evaluated patients, Dr. Jotte said. The most common grade 3/4 adverse event in either arm was fatigue, occurring in 16% of 63 erlotinib/placebo patients and in 20% of 65 erlotinib/entinostat patients. Fatal adverse events occurred in 25.4% of erlotinib/placebo patients vs. 18.5% of erlotinib/entinostat patients, with 43% of patients in each arm discontinuing treatment because of adverse events.