In the treatment of non-small cell lung cancer (NSCLC), new challenges must be met in the era of targeted agents. These challenges include the proper selection of patients most likely to benefit from each agent, the development of approaches that will combat the redundancy within the signaling pathways that allows tumor cells to escape, the determination of optimal combinations of agents, and a more complete understanding of just how the targeted agents actually work, said Wallace Akerley, MD, Professor of Medicine and Director of Clinical Research, Huntsman Cancer Institute at the University of Utah, Salt Lake City.
“We need better treatments because the overall cure rate of NSCLC is still less than 25%, compared to 41% for other cancers,” noted Dr. Akerley, who described the innovations in treatment offered by targeted agents in NSCLC.
Multiple molecular “drivers” of tumor-specific pathways abound in solid tumors such as lung cancer. These drivers lead to redundancy within growth-signaling pathways, allowing tumor cells to escape attack. For treatment to be effective, Dr. Akerley asserted, this redundancy must be overcome. A multitargeted approach, therefore, is most attractive. Inhibitors of epidermal growth factor receptor (EGFR) and vascular endothelial growth factor (VEGF) are the classes of agents being applied so far to this “mini-ecosystem” of multiple potential targets of exploitation.
Phase II trials of the EGFR inhibitors gefitinib (Iressa) and erlotinib (Tarceva) in NSCLC were highly encouraging. “We saw response rates of 10% in the second-line setting that occurred as quickly as chemotherapy and dramatic reductions of lesions on CT scans with just one pill a day,” Dr. Akerley said.
“We thought we would get even better results when we combined these agents with chemotherapy,” he remarked, “but we got a let-down—no improvement in survival. We learned that using EGFR inhibitors with chemotherapy in unselected patients is probably not a good idea.”
However, in the BR.21 trial, which was conducted in metastatic patients progressing on chemotherapy, Shepherd et al (N Engl J Med 2005;353:123–132) showed that erlotinib significantly improved survival over best supportive care, with a 27% reduction in mortality, a median survival of 6.7 months versus 4.7 months, and a 1-year survival of 31.2% versus 21.5% (P < 0.001). “And these were patients who [normally] would have gone to hospice,” he observed.
A survival benefit was seen across all subgroups. A multivariate analysis determined that adenocarcinoma histology, non-smoking status, and Asian race were the clinical factors most strongly associated with survival. Protein expression of EGFR was also associated with response; however, it was not significantly associated with survival, nor was the number of EGFR gene copies or presence of EGFR mutations. “This was a let-down for me in terms of having a means of customizing treatment,” he commented.
EGFR positivity by fluorescence in situ hybridization (FISH) analysis was predictive of response, but the clinical factors proved to be more predictive. For predicting mortality, molecular markers lost their significance, Dr. Akerley said. Thus, it may be cost effective to use just clinical factors when assessing patients. “Perhaps EGFR-negative patients should not receive EGFR inhibitors in NSCLC,” he concluded.
EGFR inhibitors can, however, yield impressive results in properly selected patients. “How good can treatment be with a single targeted agent? Good, depending on the population,” Dr. Akerley observed. One-year survival rates of 76%–80% have been achieved in populations that are predominantly Asian, never-smokers, and/or have an EGFR mutation.
Whether an EGFR inhibitor might yield better outcomes than chemotherapy was recently explored in the phase III INTEREST (NCT00076388; Iressa NSCLC Trial Evaluating Response and Survival Against Taxotere) trial, but the results were disappointing, he noted. INTEREST evaluated gefitinib versus docetaxel (Taxotere) in NSCLC patients who progressed after platinum-based chemotherapy and found no difference in 1-year survival between the arms, at 32% and 34%, respectively (Douillard J-Y et al. Presented at the 12th World Conference on Lung Cancer; September 2–6, 2007; Seoul, South Korea). Survival was not affected by race, gender, smoking status, or mutational status. “The lack of association with mutational status, flies in the face of what we want to believe about these agents,” he added.
“Maybe we can combine targeted agents with chemotherapy in NSCLC, but only the monoclonal antibodies,” Dr. Akerley suggested, based on emerging results of the FLEX (First-Line EGFR Expressing NSCLC) trial. The study included 1,100 patients who overexpressed EGFR and had initially been treated with vinorelbine and cisplatin, with or without cetuximab (Erbitux). Improved survival has been reported with the addition of cetuximab, although no further details were available.
Agents that target the VEGF pathway—bevacizumab (Avastin) in particular—have also emerged as important therapies in NSCLC. The phase III ECOG 4599 trial found that bevacizumab added to chemotherapy (paclitaxel and carboplatin) improved survival versus chemotherapy alone in advanced or metastatic disease (Sandler A et al. N Engl J Med 2006;355:2542–2550). The odds of dying were reduced by 23% with the combination, with a median survival of 12.5 months with bevacizumab combination therapy versus 10.2 months with chemotherapy alone. One-year survival was improved to 51.9% with the combination regimen versus 43.7% for chemotherapy alone, and 2-year survival increased from 16.9% to 22.1%. Although the addition of bevacizumab increased the incidence of hemorrhage, hypertension, and proteinuria, outcomes still improved, Dr. Akerley noted.
In a subgroup analysis by age (Ramalingam SS et al. J Clin Oncol 2008;26:60–65), elderly patients (³ 70 years) benefited from bevacizumab in terms of improved response rates; however, the greater toxicity cancelled out a survival benefit for the combination. “We need to remember this in our older patients,” he commented.
The AVAiL (Avastin in Lung) study in 1,043 previously untreated patients with advanced or recurrent NSCLC also found improved response rates and longer response durations in patients treated with bevacizumab plus chemotherapy (cisplatin and gemcitabine [Gemzar]) compared with chemotherapy alone; however, survival data are not yet mature (Manegold C et al. J Clin Oncol 2007;25[18S]:LBA7514). The rate of pulmonary hemorrhage was also higher in the bevacizumab-containing arm.
Currently generating much interest are studies combining EGFR and VEGF inhibitors, such as a three-arm phase II study comparing bevacizumab plus erlotinib versus bevacizumab plus chemotherapy versus chemotherapy alone (Herbst R et al. J Clin Oncol 2007;25:4743–4750). Numerically, the best results were seen with the bevacizumab-erlotinib combination, with a response rate of 18%, a median survival of 13.7 months, and a 1-year survival of 57%. Patients receiving bevacizumab with chemotherapy (either docetaxel or pemetrexed [Alimta]) had a response rate of 13%, a median survival of 12.6 months, and a 1-year survival of 54%, whereas those given chemotherapy alone achieved a 12% response rate, a median survival of 8.6 months, and a 1-year survival of 33%. The study, however, was insufficiently powered to determine whether there were significant differences among the three groups.
In his own phase II trial, Dr. Akerley is evaluating bevacizumab plus erlotinib as initial therapy in patients with non-squamous NSCLC without hemoptysis or brain metastases. Patients also will undergo serial PET imaging to evaluate proliferation, vascularity, and metabolism as it pertains to their predictive value. In addition, the investigators will analyze gene microarrays from peripheral blood mononuclear cells to identify genes related to rash, which correlates with treatment response to erlotinib.
Other multitargeted agents under evaluation in NSCLC include sunitinib (Sutent), sorafenib (Nexavar), and vandetanib (Zactima), which are showing encouraging results in preliminary studies, he added.