Lung cancer remains the leading cause of cancer-related mortality in the United States. The predominant histologic type of lung cancer is Non-Small-Cell lung cancer (NSCLC). Approximately 30% of newly diagnosed patients with NSCLC present with stages I-IIIA disease.
Click here to earn Continuing Medical Education CreditIntroduction
Lung cancer remains the leading cause of cancer-related mortality in the United States.1 The predominant histologic type of lung cancer is Non-Small-Cell lung cancer (NSCLC).2 Approximately 30% of newly diagnosed patients with NSCLC present with stages I-IIIA disease.2 It is vitally important to properly stage patients, both systemically as well as documenting the pathologic status of the mediastinal lymph nodes. More often than not, this process involves mediastinoscopy, which remains the gold standard for evaluation of the mediastinal contents. Documentation of mediastinal lymph node involvement certainly should preclude surgery as the initial treatment.
Two randomized phase III trials in N2-positive patients (stage IIIA) failed to show any survival benefit for surgery following induction chemotherapy or chemoradiotherapy compared with chemoradiotherapy alone,3,4 thereby making the role of routine surgery in this subset of NSCLC questionable. Surgical resection remains the standard of care for patients with stages I-II NSCLC who have adequate cardiopulmonary function and are candidates for resection. There is general agreement that surgical resection should be performed in those patients who go to the operating room (presumably following a negative mediastinal evaluation) and are found to have resectable stage IIIA disease that was unsuspected previously (incidental N2 disease).
The cure rate of surgery is dependent on the pathologic stage of the patient determined at the time of resection. Table 1 summarizes historic survival data from a seminal paper published 10 years ago.5 In 1995, the NSCLC Collaborative Group reported a meta-analysis evaluating the role of cisplatin-based chemotherapy in curatively resected NSCLC.6 There was an absolute benefit of 5% in survival at 5 years in favor of adjuvant cisplatin-based chemotherapy, but it failed to achieve statistical significance (hazard ratio [HR] = 0.87; P = 0.08). After this report, the general consensus was that a true survival effect of adjuvant chemotherapy could not be ruled out and that further randomized trials were needed.7
Click to enlarge
Table 2 summarizes the five major cisplatin-based trials8-12 reported since the 1995 meta-analysis.6 The trials included slightly different patient populations and used various cisplatin-based approaches for 3-4 cycles administered generally within 60 days of resection. Only one trial8 reported on the rate of complete lymph node dissection, leaving the issue of surgical quality undefined in the other four trials. Three of the trials8,10,12 defined a specific regimen to be used, whereas two trials allowed physicians a choice of regimen.9,11 The only regimen considered to be a “modern” platinum-based regimen was cisplatin plus vinorelbine, and only two of the trials10,12 mandated treatment with this regimen. Four8,9,11,12 of the five trials allowed adjuvant thoracic radiotherapy (TRT) to be delivered following chemotherapy. Three of the trials9,10,12 showed a significant effect on survival as a result of adjuvant chemotherapy, whereas two trials did not (Table 3).8,11 In those trials that were positive for survival, the HR for the effect of adjuvant cisplatin-based chemotherapy on survival ranged from 0.69-0.87. All trials demonstrated that the delivery of adjuvant cisplatin-based regimens was difficult, with compliance rates ranging from 45%-74% in patients receiving the full prescribed course of therapy. The toxicities were largely as expected (gastrointestinal and myelosuppressive), with rates of treatment-related deaths in patients receiving adjuvant therapy ranging from 0.8%-2.8%.
Click to enlarge
Click to enlarge h3>Survival Benefit of Cisplatin-based Chemotherapy in Stages II and III Disease
Pignon and colleagues13 recently reported on the LACE (Lung Adjuvant Cisplatin Evaluation) meta-analysis, which pooled the individual patient data from these five randomized trials.8-12 A total of 4,584 patients were included, with a median follow-up of 5.1 years (range, 3.1-5.9 years). The median age of the patients included was 59 years (only 9% were > 70 years), and 80% were male. Thirty-one percent of the patients had undergone pneumonectomy, and squamous cell was the dominant histology (48% versus 39% adenocarcinoma). The main objective was to identify trial or patient characteristics associated with the benefit of adjuvant cisplatin-based therapy.
The composite HR for survival was 0.89 (95% confidence interval [CI], 0.82-0.96; P = 0.004). The absolute benefit at 5 years was 5.3%, similar to that predicted by the 1995 meta-analysis.6 The combination of cisplatin plus vinorelbine seemed to provide the greatest benefit, but this analysis was confounded by the fact that patients receiving this combination were also exposed to a greater dose intensity of cisplatin. The effect of therapy was clearly significantly positive in the node-positive (stages II and III) patients but not in the node-negative (stage I) patients. In fact, there was a suggestion of a detriment in survival in the stage IA patients, although the CIs for the survival HR overlapped 1.0. There was also a significant overall effect of cisplatin-based chemotherapy on disease-free survival (HR = 0.84; 95% CI, 0.78-0.90; P < 0.001). Along with the individual trials, the LACE meta-analysis solidified the role of adjuvant cisplatin-based chemotherapy in stages II and III (node-positive) resected NSCLC.
The optimal management of stage I NSCLC remains controversial. Very few stage IA patients were included in the trials previously discussed. The observation of a possible detrimental effect in stage IA in the LACE meta-analysis should provide a level of caution about the use of adjuvant therapy in this stage of NSCLC. The case for stage IB disease also remains controversial. In subset analyses of the three positive trials,9,10,12 there did not appear to be a benefit from treatment in stage IB patients. However, these subset analyses should be viewed with skepticism, as they were not powered to detect the small differences expected in this setting in a group of patients less likely to relapse and die (Table 1).
The Cancer and Leukemia Group B (CALGB) has reported the only trial14 in which only stage IB patients were included. This trial randomized 344 stage IB patients to undergo observation versus 4 cycles of carboplatin and paclitaxel. Reported in abstract form only, this trial was closed early due to an interim analysis showing a positive effect of treatment on survival (HR = 0.62; 90% CI, 0.44-0.89; P = 0.01). However, a recent update15 with longer patient follow-up showed a decay in the survival effect of adjuvant carboplatin and paclitaxel (HR = 0.80; 90% CI, 0.60-1.07; P = 0.1). An unplanned retrospective analysis suggested a benefit to adjuvant carboplatin and paclitaxel in patients whose tumor was > 4 cm. A final analysis of this trial is yet to come, but it has questioned the routine recommendation for adjuvant chemotherapy in the node-negative patient. Whether the most recent negative results were secondary to the use of carboplatin versus cisplatin or related to the relatively small numbers of patients with an overall better prognosis remains uncertain. This trial did show that carboplatin-based therapy may be better tolerated in this setting; 85% of patients received 4 cycles of treatment, with 55% receiving those cycles at full dose.
In all adjuvant settings, significant numbers of patients receive treatment who would have otherwise been cured by surgery alone. Likewise, many patients receive adjuvant chemotherapy (and its associated toxicities) and relapse and die despite it. Olaussen and colleagues16 analyzed patients’ surgical specimens on an international trial9 for excision repair cross-complementation group 1 (ERCC1) enzyme expression and correlated its expression with the benefit of adjuvant cisplatin-based therapy. Samples were available on 761 of the 1,867 patients enrolled (specimens were only collected from centers that enrolled at least 10 patients). ERCC1 is an enzyme responsible for the repair of DNA damage induced by cisplatin.
Several interesting observations were made. To begin, ERCC1 expression was a prognostic marker for overall survival. In patients on the observation arm, high ERCC1 expression was associated with better survival than low ERCC1 expression (5-year overall survival 46% for ERCC1-positive tumors versus 39% for ERCC1-negative tumors; adjusted HR, 0.66; 95% CI , 0.49-0.90; P = 0.009), which is consistent with other observations.17 Interestingly, ERCC1 was also predictive with regard to the benefit of adjuvant cisplatin-based chemotherapy. Patients with high ERCC1 expression (and therefore enhanced ability to repair the DNA damage induced by cisplatin) did not appear to receive any benefit from adjuvant cisplatin-based therapy (HR = 1.14; 95% CI, 0.84-1.55; P = 0.4), whereas patients with low ERCC1 expression received a substantial benefit (HR = 0.65; 95% CI, 0.50-0.85; P = 0.001; Figure 1).
Click to enlarge
In the trial reported by the National Cancer Institute of Canada (NCI-C),12 K-ras mutation was a stratification variable. Patients with K-ras mutations did not appear to derive any benefit from adjuvant cisplatin and vinorelbine (HR = 0.95; 95% CI, 0.53-1.71; P = 0.87), whereas patients with wild-type K-ras did have a significant survival advantage (HR = 0.69; 95% CI, 0.49-0.98; P = 0.03). Further analyses of ongoing or completed adjuvant trials should be performed to broaden the understanding of these markers and their usefulness in decision-making in the adjuvant chemotherapy setting.18
Along the same lines, certain patients are cured by surgery alone, and several investigators have attempted to define low-risk versus high-risk patients. Potti et al19 identified gene-expression profiles that predicted the risk of recurrence in 89 patients with early-stage NSCLC who were resected for cure. A lung metagene model that identified a low-risk and high-risk metagene profile was developed; it was more predictive of 5-year survival than clinical or pathologic stage. The metagene approach was validated in an independent group with resected NSCLC obtained from an American College of Surgeons and CALGB trial. Given the potential usefulness of this approach, CALGB investigators are planning a trial in which patients with resected stage IA will be categorized as either low risk versus high risk based on the metagene analysis. No further therapy will be given to the low-risk patients; however, the high-risk patients will be randomized to undergo either observation or adjuvant chemotherapy.
Another group of investigators20 identified a five-gene signature that distinguished good prognostic from poor prognostic groups of patients with early-stage, resected NSCLC. These interesting observations will require prospective validation, but the possibility of refining the prognosis of individual patients and influencing clinical decisions regarding the use of adjuvant chemotherapy is an exciting prospect and should be evaluated in well-designed, prospective clinical trials such as the CALGB trial previously mentioned. Although these molecular profiles may be prognostic, they may not be predictive with regard to the benefit of adjuvant cisplatin-based chemotherapy.
The adoption of systemic adjuvant chemotherapy as a new standard of care in resected NSCLC raises the issue of the role of TRT in this setting. Historically, TRT has improved locoregional control in node-positive patients but has never been demonstrated to impact overall survival.2 In fact, there is concern that the use of TRT in the adjuvant setting could have a detrimental effect on survival.21 This detrimental effect seems to be an issue in stages I-II NSCLC and not necessarily in resected stage IIIA. With improved survival from adjuvant chemotherapy, locoregional control may assume greater importance. The CALGB attempted to address this issue by randomizing resected N2 patients to receive or not to receive TRT following adjuvant carboplatin and paclitaxel.22 Unfortunately, poor accrual forced premature closure of this trial, leaving the issue unresolved. Locoregional relapse remains a significant issue in certain patients, and TRT is an option in select patients. As the concurrent use of chemoradiotherapy has not been demonstrated to be a successful strategy in the adjuvant setting,2,23 it is probably best to offer sequential therapy with chemotherapy followed by TRT as the de facto standard of care.
The paradigm of adjuvant cisplatin-based chemotherapy has been established in early-stage, curatively resected NSCLC. All patients with stages II-III resected NSCLC (node-positive) should be informed about the potential survival benefit of adjuvant chemotherapy. The case is less clear in stage I (node-negative) NSCLC, but patients with larger tumors (stage IB) were included in trials that were clearly positive with regard to survival (Table 2). The optimal platinum-based combination has not been clearly established, but the evidence favors the use of vinorelbine in combination with cisplatin.
Although we have no comparative trials in the adjuvant setting, combinations of cisplatin with other agents, such as the taxanes, gemcitabine, (Gemzar), and irinotecan (Camptosar), seem to perform at least as well as vinorelbine in the advanced-disease setting. The routine use of carboplatin-based chemotherapy should be discouraged, given the negative results of the CALGB trial and the lack of comparative data in this curative setting. However, not all postoperative patients are ideal candidates for cisplatin-based chemotherapy. Whether single-agent chemotherapy in the adjuvant setting is useful in this group is being addressed in ANITA-2, in which patients who are unsuitable for adjuvant cisplatin-based chemotherapy are being randomized to receive vinorelbine alone versus observation.7Adjuvant therapy should be administered within 60 days of resection, as was the case in the trials noted in Table 2.
Future trials should evaluate both prognostic as well as predictive factors in this setting and should be performed in the context of well-designed prospective clinical trials. In addition, two major biologic pathways have been validated as therapeutic targets in advanced NSCLC.24,25 They include the epidermal growth factor receptor (EGFR)24 and the vascular endothelial growth factor (VEGF)25 receptor pathways. Efforts to evaluate erlotinib (Tarceva; as well as gefitinib [Iressa]26) and bevacizumab (Avastin; which target the EGFR and VEGF pathways, respectively) are under way in phase III trials (Figure 2). These studies should be high-priority trials, given the potential of these agents to improve cure rates in this setting. Lastly, promising results have also been seen in a phase II trial evaluating the MAGE-3 vaccine,27 which has moved into phase III testing.
Click to enlarge
CME Post-Test and Evaluation
Activity Release Date: May 30, 2007
Activity Expiration Date: May 30, 2008
About the Activity
This activity is based on a brief article developed as part of the E-Update Series and posted on the Web. The series is geared to oncologists and addresses new treatments of cancer or modifications thereof.
This activity has been developed and approved under the direction of Beam Institute.
Activity Learning Objectives
After reading this article, participants should be able to:
(a) Compare the five major cisplatin-based trials for resected Non-Small-Cell lung cancer (NSCLC).
(b) Discuss the outcome differences with adjuvant cisplatin-based chemotherapy for stages I-III disease.
(c) Explore the use of ERCC1 expression as a prognostic marker for overall survival.
(d) Review the prognostic risk of gene-expression profiles in early-stage, resected NSCLC.
(e) Summarize the general consensus on the use of thoracic radiotherapy in the adjuvant setting.
This activity targets physicians in the fields of oncology and hematology.
This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of Beam Institute and The Oncology Group. Beam Institute is accredited by the ACCME to provide continuing medical education for physicians
Continuing Education CreditAMA PRA Category 1 Credit™
The Beam Institute designates this educational activity for a maximum of 2 AMA PRA Category 1 Credit(s)™. Physicians should only claim credit commensurate with the extent of their participation in the activity.
This activity is an independent educational activity under the direction of Beam Institute. The activity was planned and implemented in accordance with the Essential Areas and policies of the ACCME, the Ethical Opinions/Guidelines of the AMA, the FDA, the OIG, and the PhRMA Code on Interactions with Healthcare Professionals, thus assuring the highest degree of independence, fair balance, scientific rigor, and objectivity.
However, Beam Institute, the Grantor, and CMPMedica shall in no way be liable for the currency of information or for any errors, omissions, or inaccuracies in the activity. Discussions concerning drugs, dosages, and procedures may reflect the clinical experience of the author(s) or may be derived from the professional literature or other sources and may suggest uses that are investigational in nature and not approved labeling or indications. Activity participants are encouraged to refer to primary references or full prescribing information resources. The opinions and recommendations presented herein are those of the author(s) and do not necessarily reflect the views of the provider or producer.
Dr. Govindan receives research support from Bristol-Myers Sqibb, Eli Lilly, Genentech, Pfizer, Sanofi-Aventis, and serves on the speakers' bureau for Eli Lilly and Genentech. Dr. Socinski receives research support and serves on the speakers' bureau for Genentech/OSI and Sanofi-Aventis.
Copyright owned by Beam Institute, a division of CME LLC. Copyright 2007, CME LLC. All rights reserved.
We would like to hear your comments regarding this or other activities provided by Beam Institute. In addition, suggestions for future activities are welcome. Contact us at:
Director of Continuing Education
11 West 19th Street, 3rd Floor
New York, NY 10011-4280