Novel Agents in the
Treatment of Lung Cancer

New Approaches to First-Line Therapies

Anti-VEGF Antibody (Bevacizumab) in Combination with Chemotherapy in First-line Treatment of NSCLC: Common Questions Answered

Jared Weiss, MD, and Tracey Evans, MD
Abraham Cancer Center, University of Pennsylvania

Bevacizumab is a recombinant, humanized monoclonal antibody against the vascular endothelial growth factor (VEGF) that is designed to inhibit VEGF-associated angiogenesis. Several hypotheses have been advanced to explain the efficacy of bevacizumab including inhibition of neoangiogenesis,^12,13 normalization of existing tumor microvasculature leading to improved chemotherapy delivery,¹² and regression of existing microvasculature.¹³ Bevacizumab was first approved for the treatment of metastatic colon cancer, and in non–small cell lung cancer (NSCLC) was first studied in a randomized phase II trial.¹⁴ In the lung cancer study, response rates, time to progression, and overall survival were improved with the addition of bevacizumab 15 mg/kg to carboplatin plus paclitaxel. However, there were also six episodes of life-threatening hemoptysis; these events occurred primarily in patients with squamous cell histology, leading to exclusion of this histology in the phase III studies.

Phase III Trials of the Addition of Bevacizumab to Cytotoxic Chemotherapy

Several major phase III trials have evaluated the addition of bevacizumab to cytotoxic chemotherapy in advanced NSCLC. ECOG 4599¹ was the first of these trials, which established the combination of bevacizumab and cytotoxic chemotherapy as a new standard of care for eligible patients. This trial randomized 878 patients with Stage IIIb or IV NSCLC to chemotherapy alone (carboplatin and paclitaxel) or the same chemotherapy regimen with bevacizumab. Patients with squamous cell carcinoma, brain metastases, or hemoptysis and those on anticoagulation therapy were excluded. Overall survival (OS) increased from 10.3 to 12.3 months, and progression-free survival (PFS) from 4.5 to 6.2 months. Statistically significant increases in ≥ Grade 3 toxicity were minimal and included hypertension (0.7% with chemotherapy alone vs 6% with bevacizumab), hemorrhage, hemoptysis (0.2% vs 1.9%), epistaxis (0.2% vs 0.7%), grade ≥ 4 neutropenia (16.8% vs 25.5%), and febrile neutropenia (2% vs 5.2%). With ECOG 4599, for the first time in a randomized phase III trial in advanced NSCLC, the 1-year median survival was finally surpassed.

Additional studies evaluated the addition of bevacizumab to other regimens. The AVAiL¹⁵ trial randomized patients to receive cisplatin and gemcitabine alone or with bevacizumab at a dose of either 7.5 mg/kg or 15 mg/kg. Although the study was initially powered for OS, the primary endpoint was changed to PFS during accrual. Like ECOG 4599, AVAiL's exclusion criteria were broad: squamous histology, mixed adenosquamous histology if predominantly squamous, hemoptysis greater than one-half teaspoon bright-red blood per event, tumor invading or abutting major blood vessels, brain metastases or spinal cord compression, uncontrolled hypertension, thrombotic or hemorrhagic disorders in the prior 6 months, and therapeutic anticoagulation within 10 days of the first dose. Nonetheless, median PFS was improved in both the 7.5 mg/kg group (6.7 months) and 15 mg/kg group (6.5 months) as compared to the chemotherapy-alone arm (6.1 months). OS was not statistically improved with bevacizumab in this study.

A Single Phase II Study Worthy of Further Mention

Pemetrexed has recently been approved for first-line treatment of nonsquamous NSCLC, following demonstrated efficacy in a phase III trial comparing cisplatin and gemcitabine to cisplatin and pemetrexed.⁶ While OS was equivalent at 10.3 months in both arms, subgroup analysis demonstrated superior survival with the pemetrexed doublet for patients with adenocarcinoma (OS 12.6 with pemetrexed plus cisplatin compared to 10.9 months with gemcitabine plus cisplatin). Can this regimen be combined with bevacizumab, given that both pemetrexed and bevacizumab are optimal for nonsquamous patients? In a phase II study, Patel et al.¹⁶ treated 49 patients with pemetrexed, carboplatin, and bevacizumab every 3 weeks for six cycles, followed by maintenance therapy with pemetrexed and bevacizumab. The overall response rate was 55%, median OS 14.1 months, and median PFS 7.8 months. Therapy was well tolerated with 4% grade 3 neutropenia, no febrile neutropenia, 6% grade 3/4 venous thrombosis, and 8% diverticulitis.

Personalized Medicine

Personalizing cancer therapy based upon individual patient and tumor characteristics was the theme of the 2009 American Society of Clinical Oncology (ASCO) annual meeting. We are now able to target some treatments based on the molecular profile of the tumor. (See Dr. Sequist's discussion of anti-EGFR therapy in NSCLC based on EGFR mutation al status, in a later chapter). In addition, we have learned that patients whose tumors bearing K-ras mutations do not benefit from EGFR-directed antibodies in the treatment of colorectal cancer.¹⁷ Unfortunately, although we can better select patients to avoid bevacizumab-related toxicity by not treating those with squamous cell carcinoma or untreated brain metastases, selection criteria for bevacizumab efficacy within this group are lacking.

Subgroup analysis from ECOG 4599 suggested decreased benefit for both women and the elderly. The hazard ratio (HR) for bevacizumab therapy in men was reported to be 0.70 but only 0.98 for women, with confidence intervals crossing 1.0. The HR for patients less than 70 years of age was 0.71, but for patients ≥ 70 was only 0.89, with confidence intervals crossing 1.0. The incidence of grade 3-5 toxicities was also significantly higher with the addition of bevacizumab in older patients, with 87% sustaining a toxicity of grade 4 or greater in this group. There were also 7 treatment-related deaths out of 111 elderly patients receiving carboplatin/paclitaxel/bevacizumab. These data are not consistent with AVAiL, where forest plots showed at least as much benefit for older patients and for women. Thus, treatment decisions should not be made based on these post-hoc analyses. The effects of age and gender remain to be tested in prospective trials. The elderly also experienced more toxicity in the E4599 subgroup analysis; the incidence of grade 3-5 toxicities in elderly patients receiving chemo therapy plus bevacizumab was 87% compared to 70% in the under-70 patients in the bevacizumab arm (P<0.001). In addition, there were 7 treatment-related deaths out of the 111 elderly patients receiving carboplatin/paclitaxel/beva cizumab, compared with 2 treatment-related deaths in the 113 elderly patients treated with carboplatin and paclitaxel alone. The treatment-related death rate in the over-70 patients receiving chemotherapy plus bevacizumab was 6.3%, compared with 2.6% in the under-70 patients receiving chemotherapy plus bevacizumab. Thus far, forest plots for the AVAIL trial show that the elderly and female subgroups experience similar benefits from bevacizumab as the group as a whole. Therefore, it is difficult to determine with certainty, based on retrospective analyses, how the female and elderly subgroups should be treated. Ideally, prospective studies would evaluate this, though this is unlikely. Further analysis of ongoing studies such as ATLAS and the Aries observational cohort study may provide further enlightenment on this topic. For now, it is safe to say that elderly patients should be approached with caution when considering treatment with bevacizumab.

Other hypotheses were raised by ECOG 4599 that are worthy of further evaluation. Clinically, hypertension arose as a potential marker of efficacy, with an OS of 14 months in patients with hypertension compared to 11 months in patients without this side effect.¹⁸ Low intracecllular adhesion molecule level (ICAM) emerged as a potential predictive biomarker.¹⁹ Finally, multiple single nucleotide polymorphisms (SNPs), including ICAM1 T469C, emerged as predictive marker or at least had a treatment-by-genotype interaction.²⁰

Can I Treat the Patient with Thrombosis?

The AVAiL study¹⁵ described above did allow full-dose anti-coagulation therapy for patients who developed thromboses while on study. The need for full-dose anticoagulation arose in 86 patients. Although the rate of any bleeding increased with bevacizumab (39.3% in the placebo arm versus 43.8% with bevacizumab 7.5 mg/kg group and 73.1% in the 15 mg/kg group), no patient in any group experienced grade 3-5 pulmonary hemorrhage, and the rate of any grade 3-5 bleeding was low in all groups (3.6%, 6.3%, and 7.7%, respectively). Thus, the patient who develops thrombosis while on bevacizumab may continue bevacizumab treatment with careful monitoring.

Can I Treat the Patient with Brain Metastases?

ECOG 4599 and AVAiL both excluded patients with brain metastases for fear that the metastases would bleed with therapy. Since about a third of lung cancer patients develop brain metastases, a large portion of patients who could potentially benefit from therapy with bevacizumab are excluded. The safety of treating the patient with brain metastases was addressed in both the PASSPORT and ATLAS studies. PASSPORT evaluated bevacizumab in the standard first- and second-line settings following treatment of brain metastases with stereotactic radiosurgery, whole-brain radiation therapy, or surgical resection. On the ATLAS study, patients were treated with erlotinib and bevacizumab maintenance therapy following first-line platinum-based therapy plus bevacizumab. Patients with treated brain metastases were allowed on study. Of 83 patients treated with bevacizumab in the 2 trials, no CNS hemorrhages were reported during treatment.²¹

Conclusion

For appropriately selected patients with advanced NSCLC, bevacizumab offers a modest but real survival advantage. Many clinicians and patients have celebrated this survival gain, while others fear the severe—albeit rare—toxicities of therapy. Some question the substantial financial cost of this incremental gain. Ongoing studies are evaluating the role of bevacizumab in the second-line setting, in combination with EGFR-directed therapies, and in the adjuvant setting. Additional studies seek to evaluate potential biomarkers that would allow better patient selection for bevacizumab therapy.

< < PREVIOUS

NEXT > >