- Email: [email protected]
Patients Treated With Platinum-Doublet Chemotherapy for Advanced Non–Small-Cell Lung Cancer Have Inferior Outcomes If Previously Treated With Platinum-based Chemoradiation
Original Study
Patients Treated With Platinum-Doublet Chemotherapy for Advanced NoneSmall-Cell Lung Cancer Have Inferior Outcomes If Previously Treated With Platinum-based Chemoradiation Ashvin Paramanathan,1 Benjamin Solomon,1,2 Marnie Collins,2 Michael Franco,2 Sarah Kofoed,2 Heather Francis,2 David Ball,1,2 Linda Mileshkin1,2 Abstract We performed a retrospective study of NSCLC patients treated with carboplatin and gemcitabine chemotherapy for either de-novo metastatic disease or recurrent disease after platinum-based chemo-radiation and determined that outcomes are inferior in patients previously exposed to platinum during chemo-radiation. These results suggest that non platinum-based agents or targeted therapies should be considered in this group. Introduction: The standard of care for locoregionally advanced nonesmall-cell lung cancer is concurrent platinumbased chemoradiation. Many patients relapse, and subsequent systemic treatment may involve platinum-doublet chemotherapy. It is not known if prior platinum-based chemoradiation influences the response to platinum-based chemotherapy given subsequently for relapse. Therefore, we compared outcomes in these patients with those in patients without prior treatment. Methods: A retrospective study of patients who had been treated with carboplatin and gemcitabine chemotherapy for de novo metastatic disease or recurrent nonesmall-cell lung cancer after receiving platinum-based chemoradiation. The primary outcome was progression-free survival (PFS). Results: A total of 104 patients were analyzed. The median age was 63 years (range, 35-81 years), with 63 (61%) patients with newly diagnosed disease and with 41 (39%) who were previously treated. The response rate was significantly lower for those previously exposed to chemoradiation (10% vs. 29%: P ¼ .001), as was the median PFS (3.6 months vs. 5.7 months; P ¼ .002), and median overall survival (OS) (8.6 months vs. 12.1 months; P ¼ .007). Only the treatment group was a significant predictor (P ¼ .032) of PFS by univariate analysis. In univariate analysis; sex (men; P ¼ .04), histology (squamous cell; P ¼ .04), Eastern Cooperative Oncology Group Performance Status Scale (P ¼ .002), and treatment group (P ¼ .023) predicted significantly inferior OS. Multivariate analysis showed that performance status was the only significant predictor of inferior OS. Conclusion: Outcomes were inferior in patients previously exposed to platinumbased chemoradiation. An approach of stratifying such patients in future trials of chemotherapy should be adopted. Alternative options such as noneplatinum-based agents or targeted therapies should be considered in this group. Clinical Lung Cancer, Vol. 14, No. 5, 508-12 ª 2013 Elsevier Inc. All rights reserved. Keywords: Nonesmall-cell lung cancer, Chemoradiation, Platinum resistance
Introduction Dissemination of nonesmall-cell lung cancer (NSCLC) can be diagnosed either de novo in untreated patients at presentation or
after previously attempted curative treatment of locoregional disease. Regardless, the standard of care in patients whose tumors lack a targetable mutation is the same in both settings: chemotherapy,
1
Sir Peter MacCallum Department of Oncology, University of Melbourne, Parkville Division of Cancer Medicine, Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia 2
Submitted: Oct 10, 2012; Revised: Mar 4, 2013; Accepted: Mar 26, 2013; Epub: Jun 20, 2013
508
-
Clinical Lung Cancer September 2013
Address for correspondence: David Ball, MD, Division of Radiation Oncology and Cancer Imaging, Peter MacCallum Cancer Centre, Locked Bag 1, A’Beckett St, Victoria 8006, Australia. Fax: þ61 3 9656 1424; e-mail contact: [email protected]
1525-7304/$ - see frontmatter ª 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cllc.2013.03.007
which could consist of a platinum-based doublet, a nonplatinum doublet, or a nonplatinum single agent. Supporting this is the observation by Sekine et al1 that responses to chemotherapy were similar in patients diagnosed with stage IV NSCLC de novo compared with patients who had recurred after previous surgical resection, although survival was superior in the patients who had postoperative recurrence.1 However, the curative treatment of previously untreated locoregional disease now frequently involves the use of chemotherapy, usually platinum based, either as an adjuvant after surgery or given concomitantly with high-dose radiotherapy.2 In theory, patients who were initially treated with a platinum agent as part of chemoradiotherapy (CRT) and then relapse may have been left with a clonal population of malignant cells that are platinum resistant.3 This could potentially mean that patients who relapsed and who are re-treated with platinum-based chemotherapy may not achieve the same degree of benefit as those who receive platinum-based chemotherapy as their first-line treatment. In a small retrospective study of patients treated with first-line chemotherapy, rechallenge on relapse with the same chemotherapy (usually platinum based) to which the patients had originally responded resulted in a response rate of 29% and superior survival compared with a cohort of patients treated with a second-line agent (docetaxel), which suggests that, in this setting, rechallenge with the first-line agents is a reasonable strategy rather than switching to a different drug.4 Because we were unable to find comparable data in patients who relapsed after prior CRT, we undertook this retrospective study to determine if the limited chemotherapy administered during CRT had an impact on subsequent response and survival when patients were rechallenged with platinum-based chemotherapy compared with patients with no prior platinum exposure.
Materials and Methods Eligibility Criteria Sequential patients with a pathologic diagnosis of NSCLC who were treated with palliative carboplatin and gemcitabine chemotherapy for metastatic or locally recurrent NSCLC between January 1, 2002, and December 31, 2006, were included. Patients were divided into 2 groups: Group A. Patients with newly diagnosed metastatic NSCLC who had not received any previous treatment. Group B. Patients with NSCLC who were previously exposed to platinum-based chemotherapy as part of CRT. Due to small numbers and heterogeneous data, a third group of patients previously treated with radiation alone or noneplatinumbased chemoradiation were not included. The patients were generally treated with a standard protocol of carboplatin (AUC 5) day 1 and gemcitabine (1000 mg/m2) day 1 and 8 of a 21-day cycle. All patients treated with at least 1 cycle of chemotherapy were analyzed.
Evaluation All relevant medical information was obtained from the hospital medical records after approval of the project by the hospital ethics committee. The primary objective was to compare the progressionfree survival (PFS) of group A and group B from the time of
commencement of treatment with palliative carboplatin and gemcitabine chemotherapy for metastatic or locally recurrent disease. Secondary objectives were to compare the 2 groups with respect to response rate, clinical benefit rate (complete or partial response or stable disease), the time to progression, and overall survival (OS). Scans were routinely performed after every 2-3 cycles of chemotherapy, and tumor response was determined by using the Response Evaluation Criteria in Solid Tumors by computed tomography.5 PFS time was defined as the time from the commencement date of palliative carboplatin and gemcitabine to subsequent progression or death without prior progression. OS was defined as the time from the start of therapy with palliative carboplatin and gemcitabine to documented death due to any cause. Both time-to-event variables were censored by the study cutoff date (April 1, 2010).
Statistical Analysis PFS and OS were estimated by using Kaplan-Meier algorithms. Because the proportional hazards assumption was not satisfied, the Wilcoxon test for comparing the survival curves was used instead of the log-rank test. Univariate and multivariate analyses of prognostic factors were performed by using Cox regression to examine differences with respect to PFS and OS for each of the groups. The Fisher exact test was used to compare both the tumor response rate and the clinical response rate to chemotherapy between the 2 groups. Logistic regression was used to compare the clinical benefit rate for each group by adjusting for significant prognostic factors. Values of P < .05 were considered statistically significant. Statistical calculations were performed by using SPSS version 18.0 (SPSS Inc, Chicago, IL).
Results A total of 104 patients treated from January 1, 2002, to December 31, 2006, were identified, with 63 (61%) patients in group A and 41 (39%) in group B. The median age was 63 years (range, 35-81 years) with 71 (68%) men and 33 (32%) women. Baseline characteristics were similar in each group (Table 1). The median follow-up at the study censor date was 68 months (range, 6.5-108 months). In group B, a variety of chemotherapy regimens were used concomitantly with radiotherapy, which was given as a potentially curative treatment to the primary site and involved regional lymph nodes after appropriate staging, including a positron emission tomography, which included carboplatin and paclitaxel in 17 patients, single-agent carboplatin in 12 patients, cisplatin and vinorelbine in 7 patients, cisplatin and gemcitabine in 2 patients, single-agent cisplatin in 1 patient, and cisplatin and 5-fluorouracil in 1 patient. All of these regimens involved low-dose weekly chemotherapy given during CRT, with no patient who received full-dose systemic therapy. No additional chemotherapy was given before or after CRT. Sixty-seven (64%) patients received more than 3 cycles of carboplatin and gemcitabine chemotherapy. Second-line systemic therapy was received by 63.5% of group A (n ¼ 40) and 53.7% of group B (n ¼ 22). Agents included docetaxel (group A, 20.6% [n ¼ 13]; group B, 19.5% [n ¼ 8]), pemetrexed (group A, 20.6%; group B, 12.1%), and erlotinib (group A, 9.5% [n ¼ 6]; group B, 7.3% [n ¼ 3]). Of group A, 28.5% (n ¼ 18), and, of group B, 19.5% (n ¼ 8) had third-line therapy. Because epidermal growth
Clinical Lung Cancer September 2013
- 509
Platinum-Doublet Chemotherapy for NoneSmall-Cell Lung Cancer Table 1 Baseline Characteristics Baseline/prognostic factors
Group A (n [ 63), no. (%)
Group B (n [ 41), no. (%)
Men
42 (66.7)
29 (70.7)
Median age > 62.5 y
31 (49.2)
21 (51.2)
11 (17.5)
2 (4.9)
P Value
Smoking history Nonsmoker
1 .08
Ex-smoker
27 (42.9)
17 (41.5)
Smoker
20 (31.7)
20 (48.8)
Unknown
5 (7.9)
2 (4.9)
Adenocarcinoma and/or BACa
27 (42.9)
15 (36.5)
Squamous cell carcinoma
10 (15.9)
11 (26.8)
Histology
Large cell carcinoma
.83
.41
8 (12.7)
7 (17.1)
18 (28.6)
8 (19.5)
0
14 (22.2)
9 (22.0)
1
35 (55.6)
22 (53.7)
2
14 (22.2)
10 (24.4)
Metastatic disease
56 (88.9)
35 (85.4)
Colinet comorbidity score < 9b
41 (70.7)
27 (69.2)
.94
Weight loss > 10% body weight
18 (28.6)
12 (29.3)
.88
Visceral only
39 (61.9%)
29 (70.7%)
Bony metastases only
13 (20.6%)
6 (14.6%)
Visceral and bone
11 (17.5%)
6 (14.6%)
Undifferentiated NSCLC ECOG PS
.96
Sites of metastases
.76
.64
Abbreviations: BAC ¼ broncho-alveolar carcinoma; ECOG PS ¼ Eastern Cooperative Oncology Group Performance Status Scale; NSCLC ¼ nonesmall-cell lung cancer. a Only one patient from each group had BAC. b From Ref. 20.
factor receptor (EGFR) inhibitors were not routinely available at the time that these patients were treated, very few patients had testing for EGFR mutations; however, EGFR mutations were identified in 2 patients in group A and in 1 patient in group B.
Survival Outcomes
510
-
Both PFS and OS were significantly worse for patients in group B in univariate analyses. The median PFS for group B was 3.6 months: (95% CI, 2.6-5.0 months) compared with 5.7 months (95% CI, 4.4-6.9 months) for group A (P ¼ .002) (Figure 1). The 6-month PFS was 17.1% (95% CI, 8.7%-33.5%) for group B compared with 46.0% (95% CI, 35.2%-60.1%) for group A. After multivariate analysis of the impact of prognostic factors (Table 1) on outcome, only the treatment group was found to significantly influence PFS with a hazard ratio of 1.57 (95% CI, 1.05-2.34) for group B relative to group A (P ¼ .03). The median OS was 8.6 months (95% CI, 4.8-10.6 months) for group B compared with 12.1 months (95% CI, 9.3-14.3 months) for group A (P ¼ .007) (Figure 2). The 6-month OS was 53.7% (95% CI, 40.4%-71.3%) for group B compared with 82.5% (95% CI, 73.7%-92.5%) for group A. The 12-month OS was 26.8% (95% CI, 16.2%-44.5%) for group B compared with 50.0% (95% CI, 39.0%-64.2%) for group A. Univariate analysis of prognostic factors showed that male sex (P ¼ .04), current smoking status (P ¼ .01), ECOG PS (Eastern Cooperative Oncology Group
Clinical Lung Cancer September 2013
Performance Status Scale) of 2 (P ¼ .002), and squamous cell histology (P ¼ .04) were statistically significant predictors of inferior OS. Neither initial tumor stage or site of progression (local vs. distant) was a significant factor for response or survival. After accounting for these variables in a multivariate analysis, only the ECOG PS score (P ¼ .004) significantly influenced the OS. The hazard ratio for the treatment (group B relative to group A) was 1.42 (95% CI, 0.91-2.21; P ¼ .12). For group B, the median OS from the time of commencement of carboplatin and gemcitabine chemotherapy increased as the time between prior treatment and disease relapse lengthened. The median OS for patients relapsing within 6 months (n ¼ 15) was 4.3 months (95% CI, 3.3-11.4 months), between 6 and 12 months (n ¼ 16) was 8.9 months (95% CI, 3.4-21.8 months), between 12 and 24 months (n ¼ 4) was 10.1 months (95% CI, 4.8 months to not reached), and beyond 24 months (n ¼ 5) was 14.1 months (95% CI, 4.7 months to not reached). However, due to small numbers, these were not significantly different. Of note, the median OS from the time of diagnosis for group B was 20.3 months (95% CI, 15.3-24.7 months).
Response to Treatment The response rates were significantly lower for group B, with only 9.8% (n ¼ 4) of patients achieving a response compared with 28.6% (n ¼ 18) of the patients in group A (P < .001). The clinical
Ashvin Paramanathan et al Figure 1 Progression-free Survival by Treatment Group
Figure 2 Overall Survival by Treatment Group
benefit rate (response plus stable disease) was also significantly lower for group B (41.5%, n ¼ 17) as opposed to group A (77.8%, n ¼ 49) (P < .001). Logistic regression, including baseline prognostic factors, showed only that the treatment group was a significant predictor of response (odds ratio [group B relative to group A] 0.20 [95% CI, 0.09-0.48]; P < .001).
patients may need re-evaluation. A recent meta-analysis showed that similar 1-year survival rates were achieved with noneplatinumbased doublets as first-line therapy.8 Recently published treatment guidelines from the American Society of Clinical Oncology suggest that noneplatinum-based doublets may provide an alternative option to certain groups of patients.9 Given the higher toxicities experienced by patients on platinum-based doublet therapy and the similar survival rates for both platinum and noneplatinumbased doublet therapy, it would not be unreasonable to advocate a change in treatment protocol for this particular group of patients. Alternatively, these patients could be treated with a nonplatinum single agent such as pemetrexed, docetaxel, or erlotinib, depending on the clinical situation. Many previous prospective trials and retrospective studies have not sufficiently documented patients’ prior treatment history, including platinum exposure. More recent trials of first-line chemotherapy for metastatic disease have excluded patients who relapsed after prior CRT, although this is a common group of patients seen in the clinic. Hence, given the potential for treatment resistance, prior platinum history should be collected from patients enrolled in future trials of chemotherapy for metastatic disease. Subsequently, this should extend to all chemotherapeutic agents because these data may also have implications for patients exposed to platinum as part of adjuvant therapy, which was not in routine use at our center at the time of our study. The results of our study are limited by the study being retrospective and involving relatively small numbers of patients. In this study, the patients who had prior platinum that contained CRT were compared with the control group, made up of patients who had newly diagnosed previously untreated metastatic NSCLC. Numerous factors beyond platinum resistance could explain the results, because the patient groups were fundamentally different. Prior exposure to radiotherapy (which may affect the tumor vasculature and impair drug delivery in the case of patients with locally recurrent disease), aggressive disease biology and length of
Discussion At present, treatment guidelines for patients who present with metastatic NSCLC and who do not harbor an EGFR mutation recommend the administration of systemic chemotherapy and discuss the evidence for a benefit in fit patients (ECOG 0 - 1) of first-line platinum-doublet chemotherapy.6 This is based on previous trials in metastatic disease, which have included both de novo patients and those who have relapsed after local therapies. Our unit protocol had been to treat all fit patients who present with metastatic disease with a platinum doublet, regardless of their treatment history. However, analysis of our results suggests that the outcome of this treatment may be impacted by prior exposure to even low doses of platinum as part of CRT. The results show significantly worse PFS and response rates in the group of patients previously treated with platinum as part of CRT compared with patients with no prior treatment. Platinum resistance is a consideration when second-line treatment for metastatic NSCLC is initiated after failure of first-line treatment.3 It is possible that previously studied molecular mechanisms of platinum resistance may also be responsible for the worse outcomes observed in the group of patients previously treated with platinumbased CRT.7 The 29% response rate observed by Nagano et al4 when patients were rechallenged on relapse with platinum-based chemotherapy does not support the platinum-resistance hypothesis, but this was a small study, and, with only 28 patients, the confidence interval for the response rate will be wide. Given our results, it is plausible to suggest that the treatment protocol for relapsed metastatic NSLC in this particular group of
Clinical Lung Cancer September 2013
- 511
Platinum-Doublet Chemotherapy for NoneSmall-Cell Lung Cancer time with disease affliction were just some of the confounding factors for which study could not control. However, the median OS from the time of diagnosis in the previously treated group of 20.3 months is consistent with what would be expected for patients treated with CRT for locally advanced disease. Further, the site of progression whether local or distant did not affect response rates or survival on univariate analysis. We were not able to conclusively determine whether increasing length of time because exposure to platinum chemotherapy was associated with longer survival, although there seemed to be a trend in this direction. In the treatment of cervical cancer, it has been shown that the shorter the time interval between completion of primary treatment with platinum-based CRT and subsequent treatment with platinum for metastatic disease, the worse the outcome. The response rates in metastatic cervical cancer are similarly low to those seen in our study if platinum-based CRT has been given fewer than 6 months before treatment.10 However, equivalent results are seen in those treated who present with de novo metastatic disease and those who develop metastatic disease more than 12 months after prior platinum-based CRT. This could relate to the more aggressive nature of relapsed disease or a potential resistance to platinum in those who relapse quickly after CRT. However, given the retrospective nature of this study, we cannot determine the causative mechanism behind these results and future prospective work is needed.
Conclusion In light of these hypothesis-generating results and the alternative option of nonplatinum therapy and targeted therapy, the treatment protocols for metastatic disease may need further clarification. Those patients who develop metastatic disease after prior treatment for locally advanced disease should not be considered the same way as those who present de novo with metastatic disease. For example, those who develop metastatic disease within 6 months of platinumbased CRT or adjuvant chemotherapy should receive treatment with a nonplatinum agent, whereas those who relapse more than 12 months after prior platinum-based chemotherapy may still gain benefit from treatment with platinum-doublet chemotherapy. Future trials are needed in the group of patients who have a relapse after CRT, such as a comparison of platinum and noneplatinumdoublet chemotherapy, to help guide clinical practice and recommendations. In addition, testing of available tumor specimens from previously performed prospective trials for indications of platinum
512
-
Clinical Lung Cancer September 2013
resistance may be able to prove unequivocally that platinum resistance is the reason for the inferior outcomes observed in this particular group of patients.
Clinical Practice Points The standard of care for loco-regionally advanced non-small lung
cancer (NSCLC) is concurrent platinum-based chemoradiation. Unfortunately many patients still relapse after this treatment and
require systemic therapy Prior exposure to platinum as part of chemoradiation may reduce
the likelihood of response to subsequent platinum therapy Those who relapse less than 12 months after platinum-based
therapy are best treated with non-platinum based chemotherapy or targeted treatment depending on the mutational status of their tumour An approach of stratifying patients for prior platinum exposure in future trials involving platinum-doublet chemotherapy should be adopted.
Disclosure The authors have stated that they have no conflicts of interest.
References 1. Sekine I, Nokihara H, Yamamoto N, et al. Comparative chemotherapeutic efficacy in non-small cell lung cancer patients with postoperative recurrence and stage IV disease. J Thorac Oncol 2009; 4:518-21. 2. Auperin A, Le Pechoux C, Pignon JP, et al. Concomitant radio-chemotherapy based on platin compounds in patients with locally advanced non-small cell lung cancer (NSCLC): a meta-analysis of individual data from 1764 patients. Ann Oncol 2006; 17:473-83. 3. Huisman C, Smit EF, Giaccone G, et al. Second-line chemotherapy in relapsing or refractory nonesmall-cell lung cancer: a review. J Clin Oncol 2000; 18:3722-30. 4. Nagano T, Kim YH, Goto K, et al. Re-challenge chemotherapy for relapsed nonesmall-cell lung cancer. Lung Cancer 2010; 69:315-8. 5. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 2000; 92:205-16. 6. Le Chevalier T, Scagliotti G, Natale R, et al. Efficacy of gemcitabine plus platinum chemotherapy compared with other platinum containing regimens in advanced nonesmall-cell lung cancer: a meta-analysis of survival outcomes. Lung Cancer 2005; 47:69-80. 7. Cosaert J, Quoix E. Platinum drugs in the treatment of nonesmall-cell lung cancer. Br J Cancer 2002; 87:825-33. 8. D’Addario G, Pintilie M, Leighl NB, et al. Platinum-based versus non-platinumbased chemotherapy in advanced nonesmall-cell lung cancer: a meta-analysis of the published literature. J Clin Oncol 2005; 23:2926-36. 9. Pfister DG, Johnson DH, Azzoli CG, et al. American Society of Clinical Oncology Treatment of Unresectable Non-Small-Cell Lung Cancer Guideline: update 2003. J Clin Oncol 2004; 22:330-53. 10. Monk BJ, Sill MW, McMeekin DS, et al. Phase III trial of four cisplatin-containing doublet combinations in stage IVb, recurrent, or persistent cervical carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2009; 27:4649-55.
Patients Treated With Platinum-Doublet Chemotherapy for Advanced NoneSmall-Cell Lung Cancer Have Inferior Outcomes If Previously Treated With Platinum-based Chemoradiation Ashvin Paramanathan,1 Benjamin Solomon,1,2 Marnie Collins,2 Michael Franco,2 Sarah Kofoed,2 Heather Francis,2 David Ball,1,2 Linda Mileshkin1,2 Abstract We performed a retrospective study of NSCLC patients treated with carboplatin and gemcitabine chemotherapy for either de-novo metastatic disease or recurrent disease after platinum-based chemo-radiation and determined that outcomes are inferior in patients previously exposed to platinum during chemo-radiation. These results suggest that non platinum-based agents or targeted therapies should be considered in this group. Introduction: The standard of care for locoregionally advanced nonesmall-cell lung cancer is concurrent platinumbased chemoradiation. Many patients relapse, and subsequent systemic treatment may involve platinum-doublet chemotherapy. It is not known if prior platinum-based chemoradiation influences the response to platinum-based chemotherapy given subsequently for relapse. Therefore, we compared outcomes in these patients with those in patients without prior treatment. Methods: A retrospective study of patients who had been treated with carboplatin and gemcitabine chemotherapy for de novo metastatic disease or recurrent nonesmall-cell lung cancer after receiving platinum-based chemoradiation. The primary outcome was progression-free survival (PFS). Results: A total of 104 patients were analyzed. The median age was 63 years (range, 35-81 years), with 63 (61%) patients with newly diagnosed disease and with 41 (39%) who were previously treated. The response rate was significantly lower for those previously exposed to chemoradiation (10% vs. 29%: P ¼ .001), as was the median PFS (3.6 months vs. 5.7 months; P ¼ .002), and median overall survival (OS) (8.6 months vs. 12.1 months; P ¼ .007). Only the treatment group was a significant predictor (P ¼ .032) of PFS by univariate analysis. In univariate analysis; sex (men; P ¼ .04), histology (squamous cell; P ¼ .04), Eastern Cooperative Oncology Group Performance Status Scale (P ¼ .002), and treatment group (P ¼ .023) predicted significantly inferior OS. Multivariate analysis showed that performance status was the only significant predictor of inferior OS. Conclusion: Outcomes were inferior in patients previously exposed to platinumbased chemoradiation. An approach of stratifying such patients in future trials of chemotherapy should be adopted. Alternative options such as noneplatinum-based agents or targeted therapies should be considered in this group. Clinical Lung Cancer, Vol. 14, No. 5, 508-12 ª 2013 Elsevier Inc. All rights reserved. Keywords: Nonesmall-cell lung cancer, Chemoradiation, Platinum resistance
Introduction Dissemination of nonesmall-cell lung cancer (NSCLC) can be diagnosed either de novo in untreated patients at presentation or
after previously attempted curative treatment of locoregional disease. Regardless, the standard of care in patients whose tumors lack a targetable mutation is the same in both settings: chemotherapy,
1
Sir Peter MacCallum Department of Oncology, University of Melbourne, Parkville Division of Cancer Medicine, Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia 2
Submitted: Oct 10, 2012; Revised: Mar 4, 2013; Accepted: Mar 26, 2013; Epub: Jun 20, 2013
508
-
Clinical Lung Cancer September 2013
Address for correspondence: David Ball, MD, Division of Radiation Oncology and Cancer Imaging, Peter MacCallum Cancer Centre, Locked Bag 1, A’Beckett St, Victoria 8006, Australia. Fax: þ61 3 9656 1424; e-mail contact: [email protected]
1525-7304/$ - see frontmatter ª 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cllc.2013.03.007
which could consist of a platinum-based doublet, a nonplatinum doublet, or a nonplatinum single agent. Supporting this is the observation by Sekine et al1 that responses to chemotherapy were similar in patients diagnosed with stage IV NSCLC de novo compared with patients who had recurred after previous surgical resection, although survival was superior in the patients who had postoperative recurrence.1 However, the curative treatment of previously untreated locoregional disease now frequently involves the use of chemotherapy, usually platinum based, either as an adjuvant after surgery or given concomitantly with high-dose radiotherapy.2 In theory, patients who were initially treated with a platinum agent as part of chemoradiotherapy (CRT) and then relapse may have been left with a clonal population of malignant cells that are platinum resistant.3 This could potentially mean that patients who relapsed and who are re-treated with platinum-based chemotherapy may not achieve the same degree of benefit as those who receive platinum-based chemotherapy as their first-line treatment. In a small retrospective study of patients treated with first-line chemotherapy, rechallenge on relapse with the same chemotherapy (usually platinum based) to which the patients had originally responded resulted in a response rate of 29% and superior survival compared with a cohort of patients treated with a second-line agent (docetaxel), which suggests that, in this setting, rechallenge with the first-line agents is a reasonable strategy rather than switching to a different drug.4 Because we were unable to find comparable data in patients who relapsed after prior CRT, we undertook this retrospective study to determine if the limited chemotherapy administered during CRT had an impact on subsequent response and survival when patients were rechallenged with platinum-based chemotherapy compared with patients with no prior platinum exposure.
Materials and Methods Eligibility Criteria Sequential patients with a pathologic diagnosis of NSCLC who were treated with palliative carboplatin and gemcitabine chemotherapy for metastatic or locally recurrent NSCLC between January 1, 2002, and December 31, 2006, were included. Patients were divided into 2 groups: Group A. Patients with newly diagnosed metastatic NSCLC who had not received any previous treatment. Group B. Patients with NSCLC who were previously exposed to platinum-based chemotherapy as part of CRT. Due to small numbers and heterogeneous data, a third group of patients previously treated with radiation alone or noneplatinumbased chemoradiation were not included. The patients were generally treated with a standard protocol of carboplatin (AUC 5) day 1 and gemcitabine (1000 mg/m2) day 1 and 8 of a 21-day cycle. All patients treated with at least 1 cycle of chemotherapy were analyzed.
Evaluation All relevant medical information was obtained from the hospital medical records after approval of the project by the hospital ethics committee. The primary objective was to compare the progressionfree survival (PFS) of group A and group B from the time of
commencement of treatment with palliative carboplatin and gemcitabine chemotherapy for metastatic or locally recurrent disease. Secondary objectives were to compare the 2 groups with respect to response rate, clinical benefit rate (complete or partial response or stable disease), the time to progression, and overall survival (OS). Scans were routinely performed after every 2-3 cycles of chemotherapy, and tumor response was determined by using the Response Evaluation Criteria in Solid Tumors by computed tomography.5 PFS time was defined as the time from the commencement date of palliative carboplatin and gemcitabine to subsequent progression or death without prior progression. OS was defined as the time from the start of therapy with palliative carboplatin and gemcitabine to documented death due to any cause. Both time-to-event variables were censored by the study cutoff date (April 1, 2010).
Statistical Analysis PFS and OS were estimated by using Kaplan-Meier algorithms. Because the proportional hazards assumption was not satisfied, the Wilcoxon test for comparing the survival curves was used instead of the log-rank test. Univariate and multivariate analyses of prognostic factors were performed by using Cox regression to examine differences with respect to PFS and OS for each of the groups. The Fisher exact test was used to compare both the tumor response rate and the clinical response rate to chemotherapy between the 2 groups. Logistic regression was used to compare the clinical benefit rate for each group by adjusting for significant prognostic factors. Values of P < .05 were considered statistically significant. Statistical calculations were performed by using SPSS version 18.0 (SPSS Inc, Chicago, IL).
Results A total of 104 patients treated from January 1, 2002, to December 31, 2006, were identified, with 63 (61%) patients in group A and 41 (39%) in group B. The median age was 63 years (range, 35-81 years) with 71 (68%) men and 33 (32%) women. Baseline characteristics were similar in each group (Table 1). The median follow-up at the study censor date was 68 months (range, 6.5-108 months). In group B, a variety of chemotherapy regimens were used concomitantly with radiotherapy, which was given as a potentially curative treatment to the primary site and involved regional lymph nodes after appropriate staging, including a positron emission tomography, which included carboplatin and paclitaxel in 17 patients, single-agent carboplatin in 12 patients, cisplatin and vinorelbine in 7 patients, cisplatin and gemcitabine in 2 patients, single-agent cisplatin in 1 patient, and cisplatin and 5-fluorouracil in 1 patient. All of these regimens involved low-dose weekly chemotherapy given during CRT, with no patient who received full-dose systemic therapy. No additional chemotherapy was given before or after CRT. Sixty-seven (64%) patients received more than 3 cycles of carboplatin and gemcitabine chemotherapy. Second-line systemic therapy was received by 63.5% of group A (n ¼ 40) and 53.7% of group B (n ¼ 22). Agents included docetaxel (group A, 20.6% [n ¼ 13]; group B, 19.5% [n ¼ 8]), pemetrexed (group A, 20.6%; group B, 12.1%), and erlotinib (group A, 9.5% [n ¼ 6]; group B, 7.3% [n ¼ 3]). Of group A, 28.5% (n ¼ 18), and, of group B, 19.5% (n ¼ 8) had third-line therapy. Because epidermal growth
Clinical Lung Cancer September 2013
- 509
Platinum-Doublet Chemotherapy for NoneSmall-Cell Lung Cancer Table 1 Baseline Characteristics Baseline/prognostic factors
Group A (n [ 63), no. (%)
Group B (n [ 41), no. (%)
Men
42 (66.7)
29 (70.7)
Median age > 62.5 y
31 (49.2)
21 (51.2)
11 (17.5)
2 (4.9)
P Value
Smoking history Nonsmoker
1 .08
Ex-smoker
27 (42.9)
17 (41.5)
Smoker
20 (31.7)
20 (48.8)
Unknown
5 (7.9)
2 (4.9)
Adenocarcinoma and/or BACa
27 (42.9)
15 (36.5)
Squamous cell carcinoma
10 (15.9)
11 (26.8)
Histology
Large cell carcinoma
.83
.41
8 (12.7)
7 (17.1)
18 (28.6)
8 (19.5)
0
14 (22.2)
9 (22.0)
1
35 (55.6)
22 (53.7)
2
14 (22.2)
10 (24.4)
Metastatic disease
56 (88.9)
35 (85.4)
Colinet comorbidity score < 9b
41 (70.7)
27 (69.2)
.94
Weight loss > 10% body weight
18 (28.6)
12 (29.3)
.88
Visceral only
39 (61.9%)
29 (70.7%)
Bony metastases only
13 (20.6%)
6 (14.6%)
Visceral and bone
11 (17.5%)
6 (14.6%)
Undifferentiated NSCLC ECOG PS
.96
Sites of metastases
.76
.64
Abbreviations: BAC ¼ broncho-alveolar carcinoma; ECOG PS ¼ Eastern Cooperative Oncology Group Performance Status Scale; NSCLC ¼ nonesmall-cell lung cancer. a Only one patient from each group had BAC. b From Ref. 20.
factor receptor (EGFR) inhibitors were not routinely available at the time that these patients were treated, very few patients had testing for EGFR mutations; however, EGFR mutations were identified in 2 patients in group A and in 1 patient in group B.
Survival Outcomes
510
-
Both PFS and OS were significantly worse for patients in group B in univariate analyses. The median PFS for group B was 3.6 months: (95% CI, 2.6-5.0 months) compared with 5.7 months (95% CI, 4.4-6.9 months) for group A (P ¼ .002) (Figure 1). The 6-month PFS was 17.1% (95% CI, 8.7%-33.5%) for group B compared with 46.0% (95% CI, 35.2%-60.1%) for group A. After multivariate analysis of the impact of prognostic factors (Table 1) on outcome, only the treatment group was found to significantly influence PFS with a hazard ratio of 1.57 (95% CI, 1.05-2.34) for group B relative to group A (P ¼ .03). The median OS was 8.6 months (95% CI, 4.8-10.6 months) for group B compared with 12.1 months (95% CI, 9.3-14.3 months) for group A (P ¼ .007) (Figure 2). The 6-month OS was 53.7% (95% CI, 40.4%-71.3%) for group B compared with 82.5% (95% CI, 73.7%-92.5%) for group A. The 12-month OS was 26.8% (95% CI, 16.2%-44.5%) for group B compared with 50.0% (95% CI, 39.0%-64.2%) for group A. Univariate analysis of prognostic factors showed that male sex (P ¼ .04), current smoking status (P ¼ .01), ECOG PS (Eastern Cooperative Oncology Group
Clinical Lung Cancer September 2013
Performance Status Scale) of 2 (P ¼ .002), and squamous cell histology (P ¼ .04) were statistically significant predictors of inferior OS. Neither initial tumor stage or site of progression (local vs. distant) was a significant factor for response or survival. After accounting for these variables in a multivariate analysis, only the ECOG PS score (P ¼ .004) significantly influenced the OS. The hazard ratio for the treatment (group B relative to group A) was 1.42 (95% CI, 0.91-2.21; P ¼ .12). For group B, the median OS from the time of commencement of carboplatin and gemcitabine chemotherapy increased as the time between prior treatment and disease relapse lengthened. The median OS for patients relapsing within 6 months (n ¼ 15) was 4.3 months (95% CI, 3.3-11.4 months), between 6 and 12 months (n ¼ 16) was 8.9 months (95% CI, 3.4-21.8 months), between 12 and 24 months (n ¼ 4) was 10.1 months (95% CI, 4.8 months to not reached), and beyond 24 months (n ¼ 5) was 14.1 months (95% CI, 4.7 months to not reached). However, due to small numbers, these were not significantly different. Of note, the median OS from the time of diagnosis for group B was 20.3 months (95% CI, 15.3-24.7 months).
Response to Treatment The response rates were significantly lower for group B, with only 9.8% (n ¼ 4) of patients achieving a response compared with 28.6% (n ¼ 18) of the patients in group A (P < .001). The clinical
Ashvin Paramanathan et al Figure 1 Progression-free Survival by Treatment Group
Figure 2 Overall Survival by Treatment Group
benefit rate (response plus stable disease) was also significantly lower for group B (41.5%, n ¼ 17) as opposed to group A (77.8%, n ¼ 49) (P < .001). Logistic regression, including baseline prognostic factors, showed only that the treatment group was a significant predictor of response (odds ratio [group B relative to group A] 0.20 [95% CI, 0.09-0.48]; P < .001).
patients may need re-evaluation. A recent meta-analysis showed that similar 1-year survival rates were achieved with noneplatinumbased doublets as first-line therapy.8 Recently published treatment guidelines from the American Society of Clinical Oncology suggest that noneplatinum-based doublets may provide an alternative option to certain groups of patients.9 Given the higher toxicities experienced by patients on platinum-based doublet therapy and the similar survival rates for both platinum and noneplatinumbased doublet therapy, it would not be unreasonable to advocate a change in treatment protocol for this particular group of patients. Alternatively, these patients could be treated with a nonplatinum single agent such as pemetrexed, docetaxel, or erlotinib, depending on the clinical situation. Many previous prospective trials and retrospective studies have not sufficiently documented patients’ prior treatment history, including platinum exposure. More recent trials of first-line chemotherapy for metastatic disease have excluded patients who relapsed after prior CRT, although this is a common group of patients seen in the clinic. Hence, given the potential for treatment resistance, prior platinum history should be collected from patients enrolled in future trials of chemotherapy for metastatic disease. Subsequently, this should extend to all chemotherapeutic agents because these data may also have implications for patients exposed to platinum as part of adjuvant therapy, which was not in routine use at our center at the time of our study. The results of our study are limited by the study being retrospective and involving relatively small numbers of patients. In this study, the patients who had prior platinum that contained CRT were compared with the control group, made up of patients who had newly diagnosed previously untreated metastatic NSCLC. Numerous factors beyond platinum resistance could explain the results, because the patient groups were fundamentally different. Prior exposure to radiotherapy (which may affect the tumor vasculature and impair drug delivery in the case of patients with locally recurrent disease), aggressive disease biology and length of
Discussion At present, treatment guidelines for patients who present with metastatic NSCLC and who do not harbor an EGFR mutation recommend the administration of systemic chemotherapy and discuss the evidence for a benefit in fit patients (ECOG 0 - 1) of first-line platinum-doublet chemotherapy.6 This is based on previous trials in metastatic disease, which have included both de novo patients and those who have relapsed after local therapies. Our unit protocol had been to treat all fit patients who present with metastatic disease with a platinum doublet, regardless of their treatment history. However, analysis of our results suggests that the outcome of this treatment may be impacted by prior exposure to even low doses of platinum as part of CRT. The results show significantly worse PFS and response rates in the group of patients previously treated with platinum as part of CRT compared with patients with no prior treatment. Platinum resistance is a consideration when second-line treatment for metastatic NSCLC is initiated after failure of first-line treatment.3 It is possible that previously studied molecular mechanisms of platinum resistance may also be responsible for the worse outcomes observed in the group of patients previously treated with platinumbased CRT.7 The 29% response rate observed by Nagano et al4 when patients were rechallenged on relapse with platinum-based chemotherapy does not support the platinum-resistance hypothesis, but this was a small study, and, with only 28 patients, the confidence interval for the response rate will be wide. Given our results, it is plausible to suggest that the treatment protocol for relapsed metastatic NSLC in this particular group of
Clinical Lung Cancer September 2013
- 511
Platinum-Doublet Chemotherapy for NoneSmall-Cell Lung Cancer time with disease affliction were just some of the confounding factors for which study could not control. However, the median OS from the time of diagnosis in the previously treated group of 20.3 months is consistent with what would be expected for patients treated with CRT for locally advanced disease. Further, the site of progression whether local or distant did not affect response rates or survival on univariate analysis. We were not able to conclusively determine whether increasing length of time because exposure to platinum chemotherapy was associated with longer survival, although there seemed to be a trend in this direction. In the treatment of cervical cancer, it has been shown that the shorter the time interval between completion of primary treatment with platinum-based CRT and subsequent treatment with platinum for metastatic disease, the worse the outcome. The response rates in metastatic cervical cancer are similarly low to those seen in our study if platinum-based CRT has been given fewer than 6 months before treatment.10 However, equivalent results are seen in those treated who present with de novo metastatic disease and those who develop metastatic disease more than 12 months after prior platinum-based CRT. This could relate to the more aggressive nature of relapsed disease or a potential resistance to platinum in those who relapse quickly after CRT. However, given the retrospective nature of this study, we cannot determine the causative mechanism behind these results and future prospective work is needed.
Conclusion In light of these hypothesis-generating results and the alternative option of nonplatinum therapy and targeted therapy, the treatment protocols for metastatic disease may need further clarification. Those patients who develop metastatic disease after prior treatment for locally advanced disease should not be considered the same way as those who present de novo with metastatic disease. For example, those who develop metastatic disease within 6 months of platinumbased CRT or adjuvant chemotherapy should receive treatment with a nonplatinum agent, whereas those who relapse more than 12 months after prior platinum-based chemotherapy may still gain benefit from treatment with platinum-doublet chemotherapy. Future trials are needed in the group of patients who have a relapse after CRT, such as a comparison of platinum and noneplatinumdoublet chemotherapy, to help guide clinical practice and recommendations. In addition, testing of available tumor specimens from previously performed prospective trials for indications of platinum
512
-
Clinical Lung Cancer September 2013
resistance may be able to prove unequivocally that platinum resistance is the reason for the inferior outcomes observed in this particular group of patients.
Clinical Practice Points The standard of care for loco-regionally advanced non-small lung
cancer (NSCLC) is concurrent platinum-based chemoradiation. Unfortunately many patients still relapse after this treatment and
require systemic therapy Prior exposure to platinum as part of chemoradiation may reduce
the likelihood of response to subsequent platinum therapy Those who relapse less than 12 months after platinum-based
therapy are best treated with non-platinum based chemotherapy or targeted treatment depending on the mutational status of their tumour An approach of stratifying patients for prior platinum exposure in future trials involving platinum-doublet chemotherapy should be adopted.
Disclosure The authors have stated that they have no conflicts of interest.
References 1. Sekine I, Nokihara H, Yamamoto N, et al. Comparative chemotherapeutic efficacy in non-small cell lung cancer patients with postoperative recurrence and stage IV disease. J Thorac Oncol 2009; 4:518-21. 2. Auperin A, Le Pechoux C, Pignon JP, et al. Concomitant radio-chemotherapy based on platin compounds in patients with locally advanced non-small cell lung cancer (NSCLC): a meta-analysis of individual data from 1764 patients. Ann Oncol 2006; 17:473-83. 3. Huisman C, Smit EF, Giaccone G, et al. Second-line chemotherapy in relapsing or refractory nonesmall-cell lung cancer: a review. J Clin Oncol 2000; 18:3722-30. 4. Nagano T, Kim YH, Goto K, et al. Re-challenge chemotherapy for relapsed nonesmall-cell lung cancer. Lung Cancer 2010; 69:315-8. 5. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 2000; 92:205-16. 6. Le Chevalier T, Scagliotti G, Natale R, et al. Efficacy of gemcitabine plus platinum chemotherapy compared with other platinum containing regimens in advanced nonesmall-cell lung cancer: a meta-analysis of survival outcomes. Lung Cancer 2005; 47:69-80. 7. Cosaert J, Quoix E. Platinum drugs in the treatment of nonesmall-cell lung cancer. Br J Cancer 2002; 87:825-33. 8. D’Addario G, Pintilie M, Leighl NB, et al. Platinum-based versus non-platinumbased chemotherapy in advanced nonesmall-cell lung cancer: a meta-analysis of the published literature. J Clin Oncol 2005; 23:2926-36. 9. Pfister DG, Johnson DH, Azzoli CG, et al. American Society of Clinical Oncology Treatment of Unresectable Non-Small-Cell Lung Cancer Guideline: update 2003. J Clin Oncol 2004; 22:330-53. 10. Monk BJ, Sill MW, McMeekin DS, et al. Phase III trial of four cisplatin-containing doublet combinations in stage IVb, recurrent, or persistent cervical carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2009; 27:4649-55.