|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 24 (August 20), 2006: pp. 3823-3830 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.05.3181 Early Compared With Late Radiotherapy in Combined Modality Treatment for Limited Disease Small-Cell Lung Cancer: A London Lung Cancer Group Multicenter Randomized Clinical Trial and Meta-Analysis
From the University College Hospitals Trust; Cancer Research UK and University College London Cancer Trials Centre; St Bartholomew's Hospital; Guy's and St Thomas' National Health Services (NHS) Trust, London; Southend Hospital NHS Trust, Southend; Cookridge Hospital, Leeds; Addenbrookes NHS Trust, Cambridge; and Essex County Hospital NHS Trust, Essex, United Kingdom Address reprint requests to Allan K. Hackshaw, MSc, Cancer Research UK and University College London Cancer Trials Centre, Stephenson House, 158-160 N Gower St, London NW1 2ND, United Kingdom; e-mail: ah{at}ctc.ucl.ac.uk
PURPOSE: To replicate an earlier National Cancer Institute of Canada (NCIC) trial that examined the effect on survival of the timing of thoracic radiotherapy (TRT) in patients with limited disease small-cell lung cancer (SCLC). PATIENTS AND METHODS: Patients received three cycles of cyclophosphamide, doxorubicin, and vincristine alternating with three cycles of etoposide and cisplatin. Three hundred twenty five chemotherapy- and radiotherapy-naïve patients were randomly assigned to either early TRT administered concurrently in the second cycle or late TRT administered concurrently with the sixth cycle; the dose was 40 Gy in 15 fractions over 3 weeks. RESULTS: TRT was received by 92% and 82% of patients in the early and late arms, respectively (P = .01). Sixty-nine percent of patients in the early arm received all six courses of chemotherapy compared with 80% in the late arm (P = .003). There was no evidence of a survival difference; median overall survival time was 13.7 and 15.1 months in the early and late arms, respectively (P = .23). In a meta-analysis of all eight trials that compared early and late TRT, there were three in which the proportion of patients who completed their planned chemotherapy was similar between the TRT arms (hazard ratio [HR] = 0.73; 95% CI, 0.62 to 0.86) and five in which proportionally fewer patients in the early TRT arm completed their chemotherapy (HR = 1.06; 95% CI, 0.97 to 1.17). CONCLUSION: This study failed to show a survival advantage for early TRT with chemotherapy in limited-stage SCLC, unlike the NCIC trial. However, the results of a meta-analysis suggest that it is essential to ensure that the delivery of chemotherapy is optimal when administered with early TRT.
The first-line treatment for small-cell lung cancer (SCLC) is combination chemotherapy. A meta-analysis of randomized controlled trials showed a small but significant advantage for the combination of chemotherapy and thoracic radiotherapy (TRT) compared with chemotherapy alone.1 Consequently, radiotherapy has become integral to the treatment of all patients with limited-disease SCLC. The timing of TRT has been the subject of seven randomized trials.2-8 Some reported a survival benefit for early TRT,3,5,8 whereas others did not.2,4,6,7 In 1993, the National Cancer Institute of Canada (NCIC) Clinical Trials Group published their results on 308 patients with limited-disease SCLC.3 All patients received a maximum of six cycles of chemotherapy and were randomly assigned to receive 40 Gy of radiotherapy to the primary tumor site coincident with the second course of chemotherapy (week 3) or with the final sixth course at week 15. Although overall response rates did not differ between the two trial arms, progression-free survival and overall survival were superior in the early TRT arm. This was the first trial to report positive results. The first published trial by Perry et al2 showed no advantage for early versus late TRT. Therefore, the London Lung Cancer Group proceeded to replicate the NCIC trial, as closely as possible, in the United Kingdom.
The current trial used a similar protocol to the NCIC trial,3 particularly the same chemotherapy and radiotherapy regimens. Unlike the NCIC protocol, routine pulmonary function tests and bone marrow sampling were deemed impractical and not routinely performed before random assignment. Brain scans were only carried out in centers able to perform them within the short time constraint before entry onto the trial. All relevant local research ethics committee approvals were obtained, and all patients provided written informed consent.
Inclusion Criteria
Exclusion Criteria
Staging Investigations and Other Assessments During treatment, patients had a full blood profile at the time of the nadir and every 3 weeks, together with a chest radiograph. Brain, bone, and liver scans were repeated whenever clinically indicated. After completion of treatment, blood profiles and the chest radiograph were performed monthly, with other scans as indicated.
Interventions TRT. Patients were randomly assigned to early TRT administered concurrently with the first cycle of EP (week 3) or to late TRT administered concurrently with the sixth cycle of chemotherapy (ie, third cycle of EP; week 15), as shown in Figure 1. The third cycle of chemotherapy (cyclophosphamide, doxorubicin, and vincristine) in the early TRT arm was delayed for 1 week to allow patients to recover from the effects of TRT and chemotherapy, as in the NCIC trial.3
TRT consisted of 40 Gy in 15 fractions over 3 weeks using cobalt-60 or a linear accelerator. The radiation began on day 1 of the first or third course of EP (ie, week 3 or 15) provided there was no evidence of progressive disease. The technique used was anterior and parallel-opposed fields with shielding of uninvolved lung. The thoracic spine was shielded to minimize the dose to the spinal cord to 35 Gy. The field size, which was based on the prechemotherapy tumor, was to be planned to encompass the primary tumor with a minimum 2-cm margin plus the entire mediastinum, with the supraclavicular lymph nodes if they were thought to be involved. Radiotherapy was continued regardless of the neutrophil count unless there was severe toxicity. Prophylactic cotrimoxazole (2 tablets bid) was administered from day 1 of each cycle of chemotherapy in which the patient received concomitant radiotherapy until the beginning of the next cycle Prophylactic cranial irradiation. Prophylactic cranial irradiation (25 Gy in 10 fractions over 2 weeks) was administered to responding patients who had a negative CT brain scan after completion of TRT and all chemotherapy. Parallel opposing 20 x 17–cm fields were used, with a cobalt-60 or a linear accelerator. The whole brain was irradiated (with the inferior border following a line drawn to avoid the eyes), including the temporal fossae and the intracranial portion of the cranial nerves. Treatment began on approximately day 8 of the third cycle of EP in the early TRT group and 2 weeks after the end of TRT in the late group.
Random Assignment
Outcome Measures Tumor response criteria. Response was assessed according to the WHO criteria and was evaluated before each cycle of chemotherapy by chest x-ray and at monthly intervals after all protocol treatment had been completed. Toxicity. For each chemotherapy cycle, the occurrence and severity of the following toxicities were reported: nausea and vomiting, mucositis, hematuria, infection, hemoptysis, neuropathy, and pain. Organ-specific toxicity was measured for renal, hepatic, cardiac, and GI functioning. The full blood cell counts obtained during both chemotherapy and follow-up provided data on anemia, leukopenia, thrombocytopenia, and neutropenia, which were categorized into a toxicity score using WHO criteria.10 Data were also collected on the postradiotherapy occurrence of pneumonitis, myelitis, esophagitis, dermatitis, skin itching or burning, and any other event.
Sample Size
Three hundred twenty-five patients from 25 United Kingdom centers were enrolled onto the trial between 1993 and 1999. One hundred fifty-nine patients were randomly assigned to early TRT, and 166 were assigned to late TRT. The baseline characteristics are listed in Table 1.
Treatment Delivery Table 2 shows that 69% of patients (110 of 159 patients) in the early TRT arm received all six courses of chemotherapy compared with 80% of patients (132 of 166 patients) in the late TRT arm (P = .03). This was a result of fewer patients reaching six cycles. Reasons for not completing all six cycles are listed in Table 3. More patients were withdrawn as a result of toxicity in the early TRT arm or because they were judged clinically unfit to proceed. The proportion of patients experiencing a delay or reduction in chemotherapy did not materially differ between the two arms. Over all six cycles, the percentages of patients who experienced at least one treatment delay were 45% (early TRT) and 44% (late TRT); the percentages of patients who experienced at least one treatment reduction were 48% (early TRT) and 44% (late TRT); and the percentages of patients who experienced at least one treatment delay with a reduction were 18% (early TRT) and 19% (late TRT).
Table 4 lists the mean percent dose delivered for each chemotherapy drug; the doses did not differ between the arms except for etoposide and cisplatin. This is expected because these drugs were administered in cycle 6, and more patients had withdrawn from chemotherapy by cycle 6 in the early TRT arm as a result of more frequent disease progression and toxicity (Table 2).
TRT was completed according to protocol by 92% of patients (147 of 159 patients) in the early TRT arm and 82% of patients (136 of 166 patients) in the late TRT arm (P = .01). A similar proportion experienced an interruption in radiotherapy delivery (16% in the early TRT arm and 14% in the late TRT arm). Only 11 patients, all in the late TRT arm, had a delay of more than 15 days from the start of cycle 6 and the beginning of TRT. In the early and late TRT arms, the reasons for not receiving TRT were progressive disease (2% and 8%, respectively) and chemotherapy-related toxicity (1% and 4%, respectively). Death not due to SCLC occurred in 2% of early TRT patients and 4% of late TRT patients. PCI was administered to 60% of early TRT patients (96 of 159 patients) and 71% of late TRT patients (118 of 166 patients) P = .04. There was an excess of withdrawals from treatment in the early arm (18 v seven patients in the early and late arms, respectively).
Survival
Tumor Response Patients had to have received at least two cycles of chemotherapy to be eligible for evaluation of response. Table 5 lists the best overall tumor response rates achieved at any time during the whole treatment period of chemotherapy and TRT; they were similar between the two arms.
Patterns of Relapse Unlike the NCIC study, CT brain scans were not performed on all patients at entry and were performed in only 45% and 49% of patients in the early and late TRT arms, respectively (Table 6). Table 6 lists the incidence of brain relapse in the trial arms, according to whether patients had a brain scan or not. There was no evidence for a difference in the proportion of patients who experienced a brain relapse (24% v 17% in early v late TRT arms, respectively; P = .12), and this observation was consistent whether a brain scan had or had not been performed at entry (Table 6).
There was no evidence of a difference according to site of relapse between the TRT arms. However, fewer patients experienced relapse in the chest in the early TRT arm compared with the late TRT arm (26%, 41 of 159 patients v 37%, 61 of 166 patients, respectively; P = .03). There was also evidence that fewer patients experienced relapse in the spine in the early TRT arm (2%, four of 159 patients) compared with the late TRT arm (8%, 14 of 166 patients; P = .02).
Toxicity
The current trial showed no evidence of a difference in survival between patients who received early or late TRT. This is in contrast to the findings of the NCIC trial,3 which reported a significant advantage of early TRT, despite both trials using a similar protocol. Table 8 lists selected data from the two trials. The main difference in survival between the two trials occurred in the early TRT arms, with a median survival time of 13.7 months in our trial compared with 21.2 months in the NCIC trial. The results for survival in the late TRT arm were similar between the two trials. Therefore, we attempted to identify reasons for the difference in the early TRT arm.
Neither study insisted on thoracic staging by CT scan because this was not routinely available at the time the Canadian study began in 1985 or in the United Kingdom in 1993. Therefore, radiotherapy planning was based mainly on the chest radiograph. The overall response rates to the combined therapy were also similar in the two trials. There was no difference between the two trials for delivering the radiotherapy, with 91% of patients in the early TRT arm and 81% of patients in the late TRT arm receiving TRT as planned in the current trial compared with 96% and 87% of patients in the early and late arms, respectively, receiving TRT as planned in the NCIC trial. The main difference between the two trials seems to be associated with the ability to deliver chemotherapy as intended to the early TRT arm. In the NCIC trial, 83% of patients randomly assigned to this arm received all six planned courses compared with 69% of patients in our trial; this difference was statistically significant (P = .003). For the late TRT arm, the proportions of patients receiving all six cycles were similar for both trials (84% in the NCIC trial and 80% in our trial). The reasons for stopping chemotherapy early in our trial are listed in Table 3. A few more patients experienced nonhematologic toxicity or were too unwell to continue, but there were no clear differences between the two arms. It could be that, in our study, there was more caution as a result of inexperience in the simultaneous administration of chemotherapy and radiotherapy because this approach was relatively new in the United Kingdom at the time of the study. The difference in the ability to deliver the intended chemotherapy between our trial and the NCIC trial in the early TRT arm and the effect on survival prompted us to determine whether this observation was present in all trials. We examined the HR from the six other published trials2,4-8 (online only Appendix A), together with the proportion of patients who completed chemotherapy in each TRT arm. Table 9 lists the results. These eight trials can be separated into the following two predefined groups: trials in which a similar proportion of patients in the early and late TRT arms received their intended chemotherapy and trials in which the proportion of patients completing all cycles of chemotherapy was noticeably less in the early versus late TRT arm. The similarity between the effects on survival (difference in median survival and HR) in each group is striking.
The pooled HR from all eight trials is 0.96 (95% CI, 0.84 to 1.10; P = .58), indicating no survival advantage of early over late TRT. However, a test for heterogeneity was statistically significant (P = .002), indicating that the HR estimates differ from the overall estimate. To examine the effect of specified subgroups, a forest plot was derived to show visually how the estimate in each subgroup differs from the pooled estimate12 (Fig 4). The following four specific characteristics were examined: the proportions of patients in the early and late arms who received their intended chemotherapy, the proportion of patients who received PCI, the use of hyperfractionated TRT, and whether the chemotherapy regimen included platinum agents. There was an indication from a previous meta-analysis by Fried et al13 that the benefit of early TRT was associated with using hyperfractionated radiotherapy or platinum-based chemotherapy. Figure 4 shows that the only noticeable departure from the pooled HR is associated with trials in which a similar proportion of patients completed chemotherapy. The association with PCI was not clear. Two trials did not offer patients PCI, and among the four trials in which proportionally more patients in the early TRT arm received PCI,3,5,6,8 three reported a survival advantage, and the other did not.6 Therefore, the effect of PCI on the comparison of early versus late TRT is uncertain.
Among the three published trials in which the early and late TRT arms had similar proportions of patients completing their chemotherapy, patients administered early TRT lived, on average, 6.5 months longer, with an HR of 0.73 (95% CI, 0.62 to 0.86; P < .001). In the five trials in which noticeably fewer patients in the early arm completed their chemotherapy, patients administered early TRT lived, on average, 1.4 months less, with an HR of 1.07 (95% CI, 0.97 to 1.17; P = .17). In the meta-analysis by Fried et al,13 the chance of surviving to 2 years was 17% greater in the early TRT group when all seven published trials were combined (ie, excluding the current trial) risk ratio = 1.17; 95% CI, 1.02 to 1.35. Our analysis in Table 9 suggests that the effect may be greater if all intended chemotherapy cycles can be delivered (risk ratio = 1.35; 95% CI, 1.13 to 1.60). We looked at the effect of early versus late TRT in the current trial when the data were restricted to those patients who had completed all six courses of chemotherapy. Among the 242 patients who received all six courses, the HR for overall survival for early versus late TRT was 0.96 (95% CI, 0.72 to 1.28) based on 192 deaths. Although the observed HR indicates no effect, the wide CI (down to 0.72) does not exclude a true beneficial effect. If the true HR were, for example, 0.75, the power to detect this is only 49% with 192 deaths. Similarly, the HR for early versus late TRT among patients who received PCI was 0.91 (95% CI, 0.67 to 1.24), so again, we were unable to make any firm conclusion about the effect of PCI. In conclusion, we were unable to replicate the results of the NCIC study and failed to show any advantage for early TRT in limited-disease SCLC. However, after considering all eight trials we believe that, within the context of the combined approach of chemotherapy and irradiation, it is essential to ensure that the delivery of chemotherapy is optimal.
The methods used to derive the results in Table 9 are as follows: The hazard ratio (HR; and 95% CI) and median overall survival in each thoracic radiotherapy (TRT) arm were taken from each published trial (Perry MC, Eaton WL, Propert KJ, et al: N Engl J Med 316:912-918, 1987; Murray N, Coy P, Pater JL, et al: J Clin Oncol 11:336-344, 1993; Gregor A, Drings P, Burghouts J, et al: J Clin Oncol 15:2840-2847, 1997; Jeremic B, Shibamoto Y, Acimovic L, et al: J Clin Oncol 15:893-900, 1997; Work E, Nielsen OS, Bentzen SM, et al: J Clin Oncol 15:3030-3037, 1997; Skarlos DV, Samantas E, Briassoulis E et al: Ann Oncol 12:1231-1238, 2001; Takada M, Fukuoka M, Kawahara M, et al: J Clin Oncol 20:3054-3060, 2002). However, there were four articles that did not report the HR (Perry MC, Eaton WL, Propert KJ, et al: N Engl J Med 316:912-918, 1987; Murray N, Coy P, Pater JL, et al: J Clin Oncol 11:336-344, 1993; Jeremic B, Shibamoto Y, Acimovic L, et al: J Clin Oncol 15:893-900, 1997; Skarlos DV, Samantas E, Briassoulis E et al: Ann Oncol 12:1231-1238, 2001). The lead author of two of these publications sent this information as a personal communication (M. Perry, personal communication. May 2005; J. Herndon, personal communication. May 2005; N. Murray, personal communication. May, 2005; Perry MC, Eaton WL, Propert KJ, et al: N Engl J Med 316:912-918, 1987; Murray N, Coy P, Pater JL, et al: J Clin Oncol 11:336-344, 1993). For the other two trials (Jeremic B, Shibamoto Y, Acimovic L, et al: J Clin Oncol 15:893-900, 1997; Skarlos DV, Samantas E, Briassoulis E et al: Ann Oncol 12:1231-1238, 2001), it was assumed that survival followed an exponential distribution, so the HR was taken as the ratio of the median survival times, and the SE was estimated using the log of the HR and the reported P value from the log-rank test as follows: (loge HR/SE) = z value associated with ( x P value). This is likely to be a reasonable assumption. For example, the HR for the NCIC trial was 0.75 (95% CI, 0.61 to 0.93) using the indirect approach, close to the result estimated directly from the data (HR = 0.73; 95% CI, 0.56 to 0.94). The risk ratio of surviving to 2 years (and 95% CI) was taken directly from the meta-analysis by Fried et al (Fried DB, Morris DE, Poole C, et al: J Clin Oncol 22:4837-4845, 2004). The HR and risk ratio were pooled using a method that allows for random variation between trials (DerSimonian R, Laird N: Meta-analysis in clinical trials. Control Clin Trials 7:177-188, 1986). Fried et al (Fried DB, Morris DE, Poole C, et al: J Clin Oncol 22:4837-4845, 2004) combined the risk ratios using the Mantel-Haenszel method, which does not allow for between-trial variation. However, the two methods produced similar results in the seven published trials; the pooled risk ratio is 1.17 using the Mantel-Haenszel method and 1.19 using the DerSimonian and Laird method.
The following clinicians and researchers helped recruit patients to the trial: Addenbrookes National Health Services (NHS) Trust, Cambridge –D. Gilligan, L. Magee; Brompton Hospital –S. Spiro, W. Burford; Broomfield Hospital, Chelmsford –D. Blainey, D. Utting; Charing Cross Hospital, London –M. Seckl, S. Stewart, B. Phillips, M. Glaser; Cheltenham General Hospital, Cheltenham –P. Jenkins, A. Ashton, M. Walker; Clatterbridge Centre for Oncology, Clatterbridge –D. Littler, T. Murdock; Cookridge Hospital, Leeds –M. Snee, J. White; Essex County Hospital NHS Trust –P. Murray, S. Heatley; Frimley Park Hospital, Surrey –R. Knight; Guys & St Thomas NHS Trust –P. Harper; Harefield Hospital, Harefield –P. Study; Harrogate District Hospital, Harrogate –A.G. Fennerty, A. Chatten; Heatherwood, Ascot –M. Smith; Leicester Royal Infirmary, Leicester –A. Benghiat, K. O'Byrne, G.D. Thomas, C. Mason, B. Savelich; Mount Vernon Hospital, Northwood –R.D. Ashford; Norfolk & Norwich University Hospital, Norwich –C. Martin, J. Beety; Queen Mary's Hospital, Sidcup –J. Prendiville; Royal Free Hospital, London –A. Jones, K. Piggott, K. Barnes; Royal Sussex County Hospital Brighton –G. Newman; Southend Hospital NHS Trust, Southend –C. Trask, A. Lamont; St Bartholomew's, London –R. Rudd, J. Steele, P. Wells, M. Evans; University College London, London –S. Spiro; Wexham Park Hospital, Wexham –J. Wiggins; Whipps Cross University Hospital, London –R. Taylor; and Whittington Hospital NHS Trust, London –S.M. Lee.
The authors indicated no potential conflicts of interest.
Presented in part at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Pignon JP, Arriagada R, Ihde DC, et al: A meta-analysis of thoracic radiotherapy for small-cell lung cancer. N Engl J Med 327: 1618-1624, 1992[Abstract] 2. Perry MC, Eaton WL, Propert KJ, et al: Chemotherapy with or without radiation therapy in limited small-cell carcinoma of the lung. N Engl J Med 316: 912-918, 1987[Abstract] 3. Murray N, Coy P, Pater JL, et al: Importance of timing for thoracic irradiation in the combined modality treatment of limited stage small-cell lung cancer. J Clin Oncol 11: 336-344, 1993 4. Gregor A, Drings P, Burghouts J, et al: Randomized trial of alternating versus sequential radiotherapy/chemotherapy in limited-disease patients with small-cell lung cancer: A European Organisation for Research and Treatment of Cancer Lung Cancer Cooperative Group study. J Clin Oncol 15: 2840-2847, 1997[Abstract] 5. Jeremic B, Shibamoto Y, Acimovic L, et al: Initial versus delayed accelerated hyperfractionated radiation therapy and concurrent chemotherapy in limited small cell lung cancer. J Clin Oncol 15: 893-900, 1997 6. Work E, Nielsen OS, Bentzen SM, et al: Randomized study of initial versus late chest irradiation combined with chemotherapy in limited-stage small-cell lung cancer. J Clin Oncol 15: 3030-3037, 1997[Abstract] 7. Skarlos DV, Samantas E, Briassoulis E, et al: Randomized comparison of early versus late hyperfractionated thoracic irradiation concurrently with chemotherapy in limited disease small-cell lung cancer: A randomized phase II study of the Hellenic Cooperative Oncology Group (HeCOG). Ann Oncol 12: 1231-1238, 2001 8. Takada M, Fukuoka M, Kawahara M, et al: Phase III study of concurrent versus sequential thoracic radiotherapy in combination with cisplatin and etoposide for limited-stage small-cell lung cancer: Results of the Japan Clinical Oncology Group Study 9104. J Clin Oncol 20: 3054-3060, 2002 9. Evans WK, Feld R, Murray N, et al: Superiority of alternating non-cross resistant chemotherapy in extensive small cell lung cancer: A multicenter, randomized clinical trial by the National Cancer Institute of Canada. Ann Intern Med 107: 451-458, 1987 10. WHO: WHO Handbook for Reporting Results of Cancer Treatment. Geneva, Switzerland, WHO, WHO offset publication 48, 1979 11. Souhami RL, Rudd R, Ruiz de Elvira M-C, et al: Randomized trial comparing weekly versus 3-week chemotherapy in small-cell lung cancer: A Cancer Research Campaign trial. J Clin Oncol 12: 1806-1813, 1994 12. Cuzick J: Forest plots and the interpretation of subgroups. Lancet 365: 1308, 2005 13. Fried DB, Morris DE, Poole C, et al: Systematic review evaluating the timing of thoracic radiation therapy in combined modality therapy for limited-stage small-cell lung cancer. J Clin Oncol 22: 4837-4845, 2004 Submitted December 22, 2005; accepted April 14, 2006.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
3,000/µL, platelet count 
