Abstract
Background: Little information is available until today regarding the effect of platinum-doublet chemotherapy with bevacizumab followed by pemetrexed as a sole maintenance treatment. Original data concerns different induction regimens plus bevacizumab with bevacizumab maintenance to progression in advanced non-squamous non-small cell lung cancer (NSCLC). Patients and Methods: Two consecutive groups of patients with advanced non-squamous NSCLC received carboplatin, vinorelbine and bevacizumab for four cycles. Group A (2010-2012) did not receive any maintenance therapy, whereas group B (2012-2013) received pemetrexed switch maintenance (500 mg/m2) every three weeks until disease progression. Results: The median progression-free survival (PFS) was 4.4 months and 7.3 months (p<0.001) and one-year survival was 42% and 52% for Group A (n=20) and B (n=22) patients, respectively. Disease control rates were 65% and 86% (p=0.104). Deep venous thrombosis and pulmonary embolism occurred in three (15%) and seven (32%) patients in group A and B, respectively. Conclusion: The inferior PFS (p<0.001) without maintenance suggests that induction treatment including bevacizumab should not be planned without subsequent maintenance treatment. Whether it is better to use pemetrexed or bevacizumab, or a combination of both, as maintenance therapy is not yet established.
- Lung cancer
- non-small cell lung cancer
- bevacizumab
- platinum-based chemotherapy
- overall survival
- progression-free survival
- pemetrexed
- maintenance treatment
Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, with the non-squamous histological sub-type accounting for 54% of cases. The incidence of lung cancer in Denmark was 78 per 100,000 in men and 68 per 100,000 in women in 2012 (1). The prognosis is generally poor, although improving, with 1-year survival of 42% and 5-year survival of 12% for all Danish cases in 2012 (1).
Standard treatment for advanced stages of NSCLC without activating mutations of genes, such as epidermal growth factor receptor (EGFR), or the fusion gene of echinoderm microtubule-associated protein-like 4 and anaplastic lymphoma kinase (ELM4–ALK), is platinum-based doublet chemotherapy. Addition of bevacizumab to such regimens has improved survival for non-squamous lung cancer (2).
Bevacizumab is a monoclonal antibody that inhibits angiogenesis through inhibition of vascular endothelial growth factor A. Bevacizumab is currently the only anti-angiogenetic drug approved for treatment of non-squamous NSCLC; this drug was originally developed and evaluated with use of bevacizumab maintenance following induction combination treatment. However, the relative impact of the maintenance part of the treatment with bevacizumab has not been firmly established (3).
Due to toxicity concerns, we initially used the bevacizumab induction regimen without any maintenance (group A). In a subsequent cohort, we used pemetrexed as a switch maintenance treatment after induction doublet chemotherapy (group B), still without bevacizumab maintenance.
Pemetrexed is an anti-metabolite that blocks the synthesis of DNA and RNA by inhibiting the synthesis of purines and pyrimidines. Pemetrexed has a documented effect as maintenance both when used as continuation and as switch maintenance (4-6).
The aim of this retrospective analysis was to explore the efficacy of doublet chemotherapy with bevacizumab but without bevacizumab maintenance (group A), investigate the effect of pemetrexed switch maintenance on progression-free survival (PFS) and overall survival (OS) in patients receiving carboplatin, vinorelbine and bevacizumab induction treatment (group B) compared to patients of group A, and also to compare the efficacy of pemetrexed switch maintenance to results in the literature on trials using bevacizumab maintenance.
Patients and Methods
Two consecutive groups of patients with advanced non-squamous NSCLC received platinum-based doublet chemotherapy plus bevacizumab as induction treatment for four courses. Selection of patients included those with tumors without invasion or close location to large vessels, no hemoptysis, and performance status (PS) 0-1. Group A (July 2010 to April 2012) had no subsequent maintenance treatment, whereas group B (May 2012 to October 2013) received switch maintenance treatment with pemetrexed. Patients had pathologically-verified advanced disease (NSCLC stage IIIb-IV) and no other cancer diagnoses. None had uncontrolled hypertension, proteinuria, hemoptysis, or tumors close to large vessels.
Both groups received 7.5 mg/kg bevacizumab together with carboplatin, at an area under the curve of 5 (AUC 5), intravenously together with vinorelbine at 60 mg/m2 orally day 1 and vinorelbine at 80 mg/m2 day 8 every three weeks for four courses. Group A did not receive any maintenance treatment after induction treatment, while group B received switch maintenance treatment with pemetrexed 500 mg/m2 every three weeks until progression, intolerable toxicity, or consent withdrawal. Switch maintenance was administered to patients without disease progression at the evaluation after four induction courses, and folic acid, vitamin B12 and prophylactic dexamethasone were administered during pemetrexed treatment.
Outcome measures on response rate (RR), disease control rate (DCR), time of progression, and toxicity in the two groups were updated in July 2014. The time of death was updated in November 2014.
Statistical analysis was performed with IBM SPSS Statistics version 19.0 (IBM, Copenhagen, Denmark) (7), using Kaplan-Meier method for analyzing survival and independent sample t-test to compare means of efficacy for the two groups.
Literature was searched using PubMed and Cochrane Library using the key words: advanced NSCLC, bevacizumab, pemetrexed, maintenance treatment, progression-free survival and overall survival.
Results
A total of 20 patients were included in group A and 22 patients in Group B. Patients' characteristics are shown in Table I. No major differences were observed between the two groups. The median lead time from diagnosis to treatment was 29 days for group A and 27 days for group B.
Four patients in group A (20%) and three in group B (14%) had brain metastases before commencement of treatment and another patient was diagnosed with brain metastases shortly after commencing treatment (total of 18% in group B). Two patients in group B (9%) had malignant spinal cord compression at the time of diagnosis.
Three patients in group A discontinued induction treatment due to vascular complications: Two had pulmonary embolism and one had deep venous thrombosis. Five patients in group B discontinued induction treatment due to vascular complications: Three had pulmonary embolism, whereas two had deep venous thrombosis. One patient in group A experienced minor bleeding episodes (epistaxis) without any need for blood transfusion, while no patients in group B experienced bleeding episodes. Patients in group B more frequently had neutropenia (32% vs. 10%) or febrile neutropenia (14% vs. 0%) Common Terminology Criteria for Adverse Events (CTCAE) of grade 3 or more than patients in group A. No toxic deaths occurred in either of the two groups. At the time of data collection, all patients in group A had died, while six patients in group B were alive.
Regarding induction treatment, the proportion of patients who completed 1 or 2 courses and 3 or 4 courses were 35% and 65% of patients in group A, and 18% and 82% in group B, respectively (p=0.216). Seventeen patients with non-progressive disease in group B started pemetrexed switch maintenance. The median number of pemetrexed switch maintenance courses was 6 (range=2-13).
A total of 14 patients in group A (70%) received second-line chemotherapy, that was pemetrexed in 12 cases, with a median of three courses (range=2-12). In group B, 17 patients received pemetrexed switch maintenance, two patients had pemetrexed as second-line treatment, two died before commencing second-line treatment, and one had a PS lower than 2 and was unable to receive further chemotherapy.
Treatment results are shown in Table II. Group A had a RR of 30% and DCR of 65% compared to a RR of 27% and a DCR of 86% in group B (p=0.592 and p=0.104, respectively).
Median PFS was 4.0 months and 7.2 months (p<0.001) and median OS 6.7 months and 15.8 months (p=0.250) for group A and B, respectively. One-year survival was 42% and 58% for group A and B, respectively, while two-year survival was 15% for both groups (Figures 1 and 2).
Discussion
This comparative retrospective study suggests an effect on PFS and a numerical, although not significant, OS benefit for patients receiving pemetrexed switch maintenance compared to no-maintenance after induction treatment with carboplatin, vinorelbine, and bevacizumab in patients with advanced non-squamous NSCLC and wild-type EGFR, ELM4 and ALK. This is in accordance with findings for switch maintenance with pemetrexed following other combinations of induction treatments. The better PFS for the maintenance group could be explained by the fact that a larger percentage received the full induction treatment. Treatment was well-tolerated, although a higher grade of hematological toxicity was observed in group B, i.e. those receiving pemetrexed switch maintenance.
The frequency of serious adverse effects such as CTCAE grade 3 or 4 bleeding during induction treatment were comparable to those of the literature, being 5% in group A and 0% in group B (8). Thromboembolic events such as pulmonary embolism and deep venous thrombosis during induction treatment occurred in 15% of cases in group A and 23% in group B, similar to the literature (9). Thus, bevacizumab in combination with bevacizumab with carboplatin and vinorelbine seems as well-tolerated as other regimens.
It is not possible, based on this retrospective study, to make direct comparison with large prospective randomized trials. But similar results with respect to efficacy with maintenance treatment have been seen in five randomized phase III trials concerning bevacizumab with or without pemetrexed as maintenance treatment, although using induction regimens other than carboplatin and vinorelbine (2, 5, 10-14, 15).
A recent meta-analysis including 4,286 patients from 12 trials suggested switch maintenance to be superior to continuation maintenance (15). Another possible advantage is suggested in a study of Ciuleanu et al. (5), showing switch maintenance with pemetrexed delayed symptoms from cancer – particularly pain (6.1 months in the pemetrexed arm vs. 4.6 months in the placebo arm, p=0.041) and hemoptysis from any cause (hazard ratio=0.58 in the pemetrexed arm, p=0.038) (5). This may support the strategy for using pemetrexed switch maintenance in the current setting. In our study, DCR measured from induction treatment was 65% in group A and 86% in group B (p=0.104), suggesting that patients receiving switch maintenance with pemetrexed had a somewhat better DCR than patients not receiving it. The difference in the DCR between our two groups might have been influenced by the fact that patients in group B had a somewhat higher, although not significantly, number of completed induction treatments (p=0.216), that may result in a better DCR.
Studies that used bevacizumab as continuation maintenance (2, 9, 12) or as combination maintenance (12) mainly report CTCAE grade 3 or more toxicity concerning: hypertension, hemopthysis, proteinuria, and thromboembolic complications (i.e. non-hematological adverse effects). This is in contrast to what is reported in randomized trials concerning pemetrexed maintenance (5, 11, 12, 14). Concerning safety, fewer drug-related deaths occurred in the setting with pemetrexed maintenance (2, 5, 10-13). No increased toxicity was seen with two drugs as maintenance treatment compared to one drug (12). Despite the fact that the RR and DCR were similar to those of literature, the PFS and OS were lower, likely due to inclusion of patients with brain metastases and malignant spinal cord compression, giving a prognostically worse patient group.
In general, switch maintenance with pemetrexed is well-tolerated (16). Our study showed a tendency for higher hematological toxicity – CTCAE grade 3-4 neutropenia and febrile neutropenia in group B receiving pemetrexed switch maintenance. This feature is seen in several studies, together with fatigue and anemia (5, 11, 12, 14). Lower toxicity grades (CTCAE grade 1-2) are also more prominent (5). No difference in number of CTCAE grade 3 or 4 events have been noted according to number of treatment courses applied in the current study, although the study of Ciuleanu et al. (5) suggests that adverse effects of pemetrexed maintenance seem to increase with longer exposure to pemetrexed (more than six cycles). This suggests that pemetrexed switch maintenance might be beneficial and well-tolerated. No treatment-related deaths occurred.
In conclusion, this comparative retrospective study among two successive groups of patients with non-squamous NSCLC suggested a beneficial effect of switch maintenance treatment on PFS (p≤0.001) also when the induction treatment was carboplatin, vinorelbine and bevacizumab which has not previously been reported. Likewise, a numerical improvement in median OS and 1-year survival rate occurred in the group receiving pemetrexed switch maintenance compared to the group without, as shown in the literature with switch maintenance treatment, although with use of induction chemotherapy regimens. The current results also support that induction chemotherapy with bevacizumab should likely be followed by a maintenance treatment. Whether this maintenance treatment should be bevacizumab alone, pemetrexed alone or a combination of both drugs remains an open question to be answered in future randomized trials.
- Received July 2, 2015.
- Revision received August 24, 2015.
- Accepted September 24, 2015.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved