genglob magazine

Integration of Novel Agents Into Treatment Regimens for Patients With Advanced Non-Small-Cell Lung Cancer

Treatment options in non-small-cell lung cancer (NSCLC) have expanded recently due to the demonstration of efficacy of targeted agents alone or in combination with existing cytotoxic chemotherapies. Research with these approaches is ongoing, and the 2006 American Society of Clinical Oncology meeting featured presentations from studies of epidermal growth factor (EGFR) inhibitors and antiangiogenic agents, as well as on the use of other novel compounds.

EGFR Inhibitors

The 2 major classes of EGFR inhibitors in clinical use are, based on their mechanisms of action, tyrosine kinase inhibitors (TKIs) and monoclonal antibodies. The former group is exemplified by gefitinib and erlotinib while the latter includes cetuximab and panitumumab.

Phase 3 trials employing the EGFR TKIs in first-line metastatic disease yielded negative results when these agents were administered concurrently with chemotherapy. Subsequent preclinical evidence suggested that sequencing chemotherapy with EGFR inhibitors may be more efficacious. Therefore, the Southwest Oncology Group performed a randomized phase 2 trial (SWOG 0342)[1] comparing concurrent chemotherapy plus cetuximab vs sequential chemotherapy followed by cetuximab in metastatic NSCLC. The plan was to select the arm with superior overall survival and subsequently compare it with chemotherapy alone in a phase 3 trial. However, the trial, which enrolled 242 patients, revealed virtually identical median progression-free (4 months in both arms) and overall (10 months in concurrent arm, 9 months in sequential arm) survival.

Based on the notion that cancer is a disease with multiple alterations, several agents are being developed that target different pathways at once. Thus, a randomized phase 2 trial evaluated a multikinase inhibitor known as ZD6474, or vanitinib, that targets both the EGFR and vascular endothelial growth factor (VEGF) receptor. This study enrolled patients who had failed platinum-containing first-line chemotherapy and compared docetaxel plus or minus ZD6474 at either 100 mg or 300 mg daily. Progression-free survival, the primary trial end point, was higher in both experimental arms (18.7 weeks in the 100-mg arm and 17 weeks in the 300-mg arm vs 12 weeks in the docetaxel-alone arm), suggesting that the addition of ZD6474 to second-line chemotherapy is efficacious. Currently, a phase 3 trial comparing docetaxel with docetaxel plus ZD6474 100 mg daily as second-line treatment for NSCLC is being conducted.

Another trial in a similar group of patients compared ZD6474 300 mg with gefitinib 250 mg in a randomized, crossover design. Progression-free survival was significantly better in the ZD6474 arm (11 vs 8.1 weeks, P = .025), meeting the trial’s primary end point. Of the 66 patients who crossed over to the alternate arm, there was 1 objective response in a patient who switched from ZD6474 to gefitinib.

In a randomized phase 2 trial, single-agent erlotinib was compared with carboplatin/paclitaxel chemotherapy in poor performance status (PS 2) patients with treatment-naive, advanced NSCLC. The primary end point was progression-free survival. Outcome measures, including response rate (12% vs 2%), progression-free survival (3.5 vs 1.9 months), and overall survival (9.7 vs 6.7 months), favored the chemotherapy arm. Overall, quality of life was equivalent between the 2 groups — compelling the authors to conclude that chemotherapy should remain the standard of care for unselected patients.

Several groups reported on attempts to select patients for EGFR TKI therapy. From prospective trials, it appears that specific mutations in EGFR exons 19 and 21 convey an exquisite sensitivity to these agents, with response rates between 60% and 95%. Of note, tumors with exon 19 mutations appear to be more sensitive compared with those that have mutations on other loci. Moreover, 1-year survival of patients harboring EGFR mutations and receiving TKI therapy appears to be over 80%. These findings were also observed in patients with bronchioloalveolar cell carcinoma, in that those with evidence of somatic EGFR mutations had an 83% observed response rate and a 22-month median overall survival.

In a similar prospective study, investigators examined the effects of gefitinib in NSCLC with respect to EGFR gene copy number, EGFR protein expression by immunohistochemistry, and activation of an EGFR downstream protein, AKT. Subjects were required to be never smokers and/or test positive for increased EGFR gene copy number or AKT activation. The overall response rate in the 42 patients enrolled thus far was 48%, with a 1-year survival of 69%. Increased EGFR gene copy number or exon 19/21 mutations were strongly associated with a positive response.

Conversely, evidence now exists to identify patients who may not derive benefit from treatment with an EGFR TKI. A subset analysis[8] of tissue samples from patients participating in the BR.21 trial comparing erlotinib with placebo in second- or third-line metastatic NSCLC indicated that those harboring a K-ras mutation have a poorer survival (hazard ratio 1.63, P = .03 in multivariate analysis) when treated with the EGFR TKI. These findings must still be viewed as preliminary, keeping in mind the retrospective nature of the analysis and the relatively small number of patient samples available for study. Nevertheless, this report, coupled with the investigators’ prior findings of lack of benefit in EGFR-negative tumors by immunohistochemistry, suggest that patients with K-ras mutations who do not express the EGFR protein are unlikely to benefit from EGFR TKI therapy.

Antiangiogenic Agents

Following the positive results of ECOG 4599, which added bevacizumab, a monoclonal antibody directed against VEGF, to doublet chemotherapy in first-line metastatic disease, a number of antiangiogenic agents have been under investigation in NSCLC. In addition, the combination of bevacizumab and erlotinib was reported to have activity in second-line metastatic NSCLC, prompting a randomized phase 2 trial evaluating bevacizumab in combination with either chemotherapy (docetaxel or pemetrexed) or erlotinib compared with chemotherapy alone in patients whose disease was refractory to first-line chemotherapy. Consistent with the first-line therapy results, the addition of bevacizumab to either erlotinib or chemotherapy was superior with respect to response rate, progression-free survival, and overall survival. The 6-month overall survival was 72.1% for bevacizumab plus chemotherapy and 78.3% for bevacizumab plus erlotinib, compared with 62.4% for chemotherapy alone. The bevacizumab/erlotinib arm was the best tolerated.

The administration of multikinase inhibitors with predominant in vitro activity against VEGF receptors is also being investigated in NSCLC. Clinical trials with 2 similar agents — sunitinib and sorafenib — were reported in patients with disease refractory to chemotherapy. Both studies allowed prior therapy with an EGFR inhibitor. Sunitinib was administered at 50 mg daily on a 4-weeks-on/2-weeks-off schedule to 63 patients enrolled in the United States and Europe. The objective response and disease control rates (responses plus stable disease) were 10% and 53%, respectively. Toxicity with sunitinib was similar to that seen in other settings, with asthenia, myalgia, and nausea being most common. In addition, 3 subjects experienced serious hemorrhages and 5 subjects (8%) experienced grade 3 hypertension.

A phase 2 trial of single-agent sorafenib, 400 mg twice a day, in a similar cohort of patients was also reported. This study enrolled 52 subjects with no objective responses reported but a disease control rate of 59%. Adverse events were typical of the agent in other diseases and included diarrhea, palmar-plantar erythrodysesthesia, and asthenia. Grade 3 hypertension occurred in 4% of subjects, and 1 patient suffered a fatal pulmonary hemorrhage. A regimen of sorafenib plus chemotherapy is currently being investigated in a randomized phase 3 trial in first-line metastatic NSCLC.

The rare but serious incidence of pulmonary hemorrhage associated with bevacizumab was investigated in a retrospective review of patient risk factors from ECOG 4599. Although only 6 cases met the criteria for early-onset hemorrhage related to bevacizumab, there appeared to be an association between these events and the presence of cavitation on pretreatment radiographs. Whether this association is real and whether it applies to all antiangiogenic agents is unknown. Nevertheless, caution should be used when these agents are considered for patients with cavitation.

Other Novel Agents

Several other novel agents have also been tested recently in NSCLC. Two randomized trials employed agents targeting the eicosanoid pathway. This pathway is implicated in several processes, including cell differentiation, and is a key mediator of inflammation. Modulation of this pathway at several points by LY293111, including inhibition of cyclooxygenase (COX)-2 and stimulation of peroxisome proliferators-activated receptor gamma, yielded promising preclinical data; thus, a randomized phase 2 trial  of the agent in combination with cisplatin/gemcitabine was undertaken in chemotherapy-naive patients with advanced NSCLC. This 3-arm trial, which administered LY293111 at 200 mg twice a day or 600 mg twice a day vs placebo, used progression-free survival as its primary end point. After enrolling 201 patients, it appeared that LY293111 was associated with a greater degree of diarrhea, especially at the 600-mg dose, but did not improve efficacy.

The Cancer and Leukemia Group B took a similar approach to inhibition of the eicosanoid pathways by administering a COX-2 inhibitor (celecoxib), a 5-lipoxygenase inhibitor (zileuton), or both in combination with carboplatin and gemcitabine in a randomized phase 2 trial. The primary end point was the rate of failure-free survival at 9 months. The agents appeared well-tolerated in combination with chemotherapy, and, although there was no difference in the primary end point, a trend toward benefit was observed in the arm administering both inhibitors. Moreover, subjects who expressed high levels of COX-2 and received celecoxib appeared to benefit compared with their low expressing counterparts.

A clinical trial administering bortezomib, a proteasome inhibitor currently approved for use in multiple myeloma patients, in combination with gemcitabine and carboplatin was reported by the Southwest Oncology Group. This phase 2 nonrandomized trial enrolled 121 patients and reported a 21% response rate, a median progression-free survival of 5 months, and a median overall survival of 11 months. Although the median survival was encouraging, the other outcome measures suggested that the addition of bortezomib is unlikely to improve efficacy in this setting.

In a phase 3 trial, investigators compared whole brain radiation alone with a combination of motexafin gadolinium and whole brain radiation therapy in 554 NSCLC patients with brain metastases. Time to neurologic progression was the primary end point. Motexafin gadolinium did not interfere with the ability to administer whole brain radiotherapy and was generally well tolerated. Time to neurologic progression trended in favor of the experimental arm (15.4 vs 10 months) but did not reach statistical significance (P = .12). Of note, it appeared that patients who began therapy within 4 weeks of diagnosis derived the greatest benefit from motexafin gadolinium; subset analysis of this group favored the experimental arm.

Conclusion

NSCLC continues to pose a treatment challenge, especially in advanced and metastatic disease. For the first time ever, targeted agents have been shown to have demonstrated efficacy in the treatment of this disease and at least one, erlotinib, is currently approved for use. It is very likely that the next few years will witness a dramatic increase in the number of novel therapies available. Two classes of agents will dominate in the near future — EGFR inhibitors and antiangiogenic agents. However, with increasing understanding of the molecular biology driving NSCLC, a greater number of agents with varied mechanisms of action will become part of individual patient management.