Erlotinib plus pazopanib shows limited activity, increased toxicity for non-small cell lung cancer
The combination of erlotinib plus pazopanib prolonged PFS compared with erlotinib plus placebo among patients with previously treated advanced non-small cell lung cancer, according to results of a phase 2 clinical trial.
However, the combination did not provide an OS benefit and resulted in more toxic effects.
“The further development of regimens combining angiogenesis and EGFR inhibitors may be appropriate in this selected subset of NSCLC patients,” David R. Spigel, MD, chief scientific officer and director of the lung cancer research program at Sarah Cannon Research Institute, and colleagues wrote.
EGFR inhibitors — used as first-line therapy for patients with NSCLC — are less toxic than standard combination chemotherapy.
Preclinical data has suggested simultaneous inhibition of angiogenesis and the EGFR pathway may have added benefit for patients with NSCLC. Single-agent pazopanib (Votrient, Novartis) — an oral multitargeted tyrosine kinase inhibitor — led to durable responses among patients with ovarian cancer and those with refractory NSCLC.
Spigel and colleagues assessed efficacy and safety of treatment with erlotinib plus pazopanib compared with erlotinib plus placebo among patients with advanced NSCLC previously treated with one or two chemotherapy regimens.
Investigators randomly assigned patients to receive daily 150 mg oral erlotinib with either 600 mg oral pazopanib (n = 127) or placebo (n = 65). Patients in both treatment groups received a median two cycles of treatment.
Patients underwent evaluation every 8 weeks until disease progression or unacceptable toxicity.
At the time of reporting, all patients had discontinued treatment, of whom 115 discontinued due to disease progression and 30 due to toxicities.
Patients treated with the combination achieved longer PFS than patients who received placebo. Although the difference met statistical significance, it only represented a change of 0.8 months (2.6 months vs. 1.8 months; HR = 0.58; 95% CI, 0.42-0.81).
Median OS was 6.9 months (95% CI, 4.9-8.5) among patients who received the combination and 7 months (95% CI, 5-9.5) among patients who received placebo. One-year survival probability was 28% for the pazopanib group and 29% for the placebo group.
Researchers observed low objective response rates in both groups. Ten percent of patients treated with erlotinib plus pazopanib demonstrated partial responses compared with 5% of patients who received erlotinib plus placebo.
The addition of pazopanib increased toxicity, which appeared consistent with the known toxicity profile, the researchers noted.
Twenty-five percent of patients in the pazopanib group experienced at least one serious adverse event compared with 5% of patients who received placebo. Toxicity led to treatment discontinuation among 21% of patients who received pazopanib and 6% of patients who received placebo.
Common adverse events that led to discontinuation included skin toxicities (n = 9), hepatoxicities (n = 6), and nausea/vomiting/diarrhea (n = 4).
A study amendment included collection of pretreatment serum specimens from a subset of 88 patients for testing with the VeriStrat assay (Biodesix) to evaluate the role of the assay in prediction outcome of treatment the with two regimens. Researchers used the predictive value of the VeriStrat score to retrospectively correlate PFS and OS among the overall population and within each treatment group.
A majority of VeriStrat scores (72%) were good which, in turn, correlated with good ECOG performance status. Researchers found no association between VeriStrat score and history of smoking, previous therapy with bevacizumab (Avastin, Genentech) or histology. Patients who underwent VeriStrat assay with good scores demonstrated better PFS and OS than those with poor scores.
Patients with a good score and received pazopanib achieved longer PFS than those in the placebo group (median, 3.6 months vs. 1.8 months; HR = 0.47), whereas patients with poor scores had a similar PFS regardless of treatment arm. OS did not differ between the treatment arms among patients with good scores, but patients with poor scores appeared to have shorter OS if they received pazopanib (2.8 months vs. 7.5 months).
Univariate analysis showed the VeriStrat score was the best predictor of PFS (P = .0009) and OS (P = .0004).
A strong justification exists for the combination of EGFR TKIs and antiangiogenesis agents, according to Kathryn Arbour, MD, fellow, and Gregory J. Riely, MD, vice chair of the clinical trials office, both of Memorial Sloan Kettering Cancer Center.
“Such agents may change the tumor vessel physiology, resulting in increased intratumoral uptake of EGFR TKIs and, thus, potentially delay the appearance of resistance cells,” Arbour and Riely wrote in an accompanying editorial.
However, this is in early stages of investigation.
Combining EGFR TKIs and immunotherapy may pose another option for patients with advanced NSCLC.
“Although single-agent immunotherapy with PD- 1/PD-L1 inhibitors has shown limited in efficacy [among] patients with EGFR-mutant lung cancers ... [it] is an approach that is currently being explored in numerous clinical trials,” Arbour and Riely wrote. “As a cautionary note, there was significant toxicity with the combination of osimertinib [Tagrisso, AstraZeneca] and durvalumab [Imfinzi, AstraZeneca], making clear that such drugs should only be combined as part of rigorous clinical trials.” – by Melinda Stevens
Disclosures: The study authors report no relevant financial disclosures. Riely reports a consultant role with Genentech/Roche; institutional clinical research support from Astellas, Infinity, Novartis, Pfizer and Takeda; and compensation from the National Comprehensive Cancer Network for chairing a review panel for grants reviewed by AstraZeneca. Arbour reports no relevant financial disclosures