The most effective treatment for early-stage (I-IIIA) non-small cell lung cancer (NSCLC) is surgical resection. Despite surgical resection, 50–60% of patients with stages I–IIIA NSCLC relapse and die from their lung cancer.^1,2 There have been attempts to reduce the risk of relapse and death from lung cancer by giving adjuvant chemotherapy to patients after surgical resection.³ This approach has been successful in patients with breast and colon cancer, where some patients with stage I–III disease are routinely given adjuvant hormonal therapy (breast cancer), chemotherapy, or both.

Small randomized trials for patients with early-stage NSCLC performed over the last 30 years have been jointly analyzed in a meta-analysis reported in 1995.³ This analysis showed a 5% survival advantage at 5 years for patients with surgically resected early-stage NSCLC treated with cisplatin-based chemotherapy compared with those on observation alone. This led to the planning and execution of multiple national and international trials that were adequately powered to look for this small benefit of an absolute increase of 5% at 5 years. The results of these trials have been reported in the last several years. The seven trials reviewed here are those that entered more than 150 patients with surgically resected stages I–IIIA NSCLC per arm. These trials are discussed in the order in which they have appeared in the literature or were presented at international meetings. The seven trials have used different types of chemotherapy, included different proportions of stages IA–IIIA NSCLC, and have different methods of reporting their results. In addition, a meta-analysis combining data from the five largest trials that used cisplatin-based chemotherapy also has been performed and reported recently.⁴

CLINICAL TRIALS

Mitomycin, Vindesine, and Cisplatin Adjuvant Trial for Patients with Resected Stage I–IIIA NSCLC

This trial is referred to as ALPI, an acronym for Adjuvant Lung Project Italy.² Patients with stage I–IIIA who successfully underwent a complete resection of their NSCLC, including mediastinal lymph nodes, were eligible for this trial. The patients were randomized to treatment with mitomycin 8 mg/m² on day 1, vindesine 3 mg/m² on days 1 and 8, and cisplatin 100 mg/m² on day 1 of a 3-week cycle or to observation. Adverse effects were monitored during the 21-day cycles, and dose adjustments were made based on the toxicities encountered. If tolerated, chemotherapy was given for three cycles. At each institution, investigators decided whether or not to administer chest radiotherapy after the completion of chemotherapy according to their institutional policy.

Cisplatin-Based Adjuvant Trial for Patients with Resected Stage I–IIIA NSCLC

This trial is referred to as IALT, an acronym for International Adjuvant Lung Cancer Trial.¹ Patients with stage I-IIIA NSCLC who successfully underwent a complete resection of their tumor were eligible for this trial. The patients were treated with 80–120 mg/m² of cisplatin given for three to four cycles of treatment in combination with etoposide, vinblastine, vinorelbine, or vindesine or were managed by observation only. Adverse effects were monitored during the 21- to 28-day cycles, and dose adjustments were made based on the toxicities encountered. The treatment was administered for three to four cycles if the patients tolerated the chemotherapy. Investigators from each institution had the option of administering chest radiotherapy after the completion of surgery plus chemotherapy for patients with N₁ or N₂ disease according to their institutional policy.

Uracil-Tegafur Adjuvant Trial for Patients with Resected Stage I Adenocarcinoma of the Lung

This trial is referred to as the UFT trial.⁵ Patients between the ages of 45 and 75 years who underwent successful resection of their stage IA and IB adenocarcinoma of the lung were eligible for this trial. The patients were randomized to be treated with the uracil-tegafur combination at a dose of 250 mg/m² (as capsules containing 100 mg tegafur and 224 mg uracil) given orally each day for 2 years or observation. Adverse effects were monitored every 3 months during the treatment, and dose adjustments were made based on the toxicities encountered. No chest radiotherapy was administered to the patients.

Cisplatin-Based Adjuvant Trial for Patients with Potentially Resectable NSCLC

This study is referred to as the BLT trial.⁶ Patients with NSCLC deemed potentially resectable or who had undergone resection were eligible for the trial. The patients could be treated before or after surgical resection with cisplatin 50–80 mg/m² given every 3 weeks for three cycles of treatment combined with one of four regimens (i.e., mitomycin and ifosfamide, mitomycin and vinblastine, vindesine, or vinorelbine) or observation. The chemotherapy was administered either before or after chest radiotherapy at the discretion of the investigators. Adverse effects were monitored during the 21- to 28-day cycles, and dose adjustments were made based on the toxicities encountered.

Vinorelbine-Cisplatin Adjuvant Trials for Patients with Resected Stage IB–II NSCLC

Two of the clinical trials used cisplatin and vinorelbine as adjuvant chemotherapy. The first is the Canadian Cooperative Group clinical trial called JBR10.⁷ Patients with stage IB–II NSCLC who successfully underwent a complete resection of their tumor were eligible for this trial. The patients were randomized to be treated with vinorelbine 25 mg/m² weekly and cisplatin 50 mg/m² on days 1 and 8 of a 4-week cycle or to observation. Adverse effects were monitored during the 28-day cycles, and dose adjustments were made based on the toxicities encountered. Treatment was administered for four cycles if the patients tolerated the chemotherapy. No chest radiotherapy was administered to the patients. The second trial to use cisplatin and vinorelbine is the ANITA (Adjuvant Navelbine International Trialist Association) study.⁸ Patients with stages IB–IIIA NSCLC who had undergone a successful complete surgical resection were eligible for this study. Patients were randomized in a 1:1 ratio to either observation or chemotherapy. The chemotherapy consisted of vinorelbine 30 mg/m² weekly and cisplatin 100 mg/m² once every 4 weeks. Patients could receive chest radiation at the discretion of the treating physician.

Paclitaxel-Carboplatin Adjuvant Trial for Patients with Resected Stage IB NSCLC

This trial is referred to by its cooperative group name, CALGB 9633.⁹ Patients with stage IB NSCLC who successfully underwent a complete resection were eligible for this trial. The patients were randomized to be treated with paclitaxel 200 mg/m² and carboplatin with an area under the curve (AUC) of 6 mg/mL/min every 3 weeks or to observation. Adverse effects were monitored during the 21-day cycles, and dose adjustments were made based on the toxicities encountered. The treatment was administered for four cycles if the patients tolerated the chemotherapy. No chest radiotherapy was administered to the patients.

RESULTS

Trial Design and Patient Characteristics in the Adjuvant Chemotherapy Trials Following Surgical Resection of NSCLC

All seven trials reported here started patient entry in 1994-1996, shortly after the results of the meta-analysis became available (Table 76-1). They completed patient accrual by 1997-2003. The Japanese UFT trial was able to accrue 979 patients limited to those with stages IA and IB adenocarcinoma of the lung in just 3 years. The four European studies were able to complete their accrual of patients with stages I–IIIA adenocarcinoma in 5 years (ALPI and IALT trials) or 6 years (BLT and ANITA trials). The studies that took the longest to accrue (7 years) were done in the United States and Canada and focused on just stage IB or stage IB plus stage II in populations that were 64% or 65% male (CALGB 9633 and JBR10, respectively). The European trials had a predominance of men (86%, 86%, 80%, and 69% for the ALPI, ANITA, IALT, and BLT, respectively), whereas the Japanese UTF trial focused on early-stage adenocarcinoma (stages IA and IB) and had a majority of women (51%). The median ages were similar in all six trials, ranging from 59 to 62 years. The Japanese UFT trial restricted its participants to ages 45–75 years, so their age range was more limited than the European and North American trials.

Treatment of the Patients Participating in the Adjuvant Chemotherapy Trials

The frequency of more extensive surgery (pneumonectomy) was reported in six of the trials and varied with differing clinical practice and the disease stages of the patients participating in the trials (see Table 76-1). The percentage of patients treated with pneumonectomy was higher in three of the European trials (25% for ALPI, 35% for IALT, and 37% for ANITA) and the Canadian trial (24% for JBR10) than in the Japanese trial (<1%) and the U.S. trial (11% for CALGB 9633). The four European trials allowed chest radiotherapy either after the adjuvant chemotherapy or after surgery for patients randomized to the no-chemotherapy arm (see Table 76-1). In ALPI, 65% of the patients treated with chemotherapy and assigned to chest radiotherapy completed their course of treatment compared with 82% who underwent observation. In IALT, 70% of the patients treated with chemotherapy and assigned to chest radiotherapy completed their course of treatment compared with 84% who did not get chemotherapy. In the BLT trial, 52 patients (14% of the total) underwent chest radiotherapy. The other three trials with earlier-stage patients (stages IA–II) did not explicitly report on the administration of chest radiotherapy.

Most of the patients completed their planned courses of chemotherapy, and deaths related to chemotherapy treatment were rare. In the chemotherapy arm of the ALPI trial, 69% of patients received the planned three courses of mitomycin, vindesine, and cisplatin (MVP). There were 10 treatment-related deaths reported in the ALPI trial, 3 in the MVP arm and 7 in the observation arm (<1% total). In the IALT trial, 74% of patients in the chemotherapy arm received at least three courses of their planned cisplatin-based therapy. In this trial there were 7 deaths related to complications of chemotherapy administration (<1%). The patients in the Japanese UFT trial were compliant with their regimen 74% of the time at 1 year and 61% of the time at 2 years. There were no treatment-related deaths reported in the UFT trial. In the BLT trial, 64% of patients received their three planned cycles of cisplatin-based therapy. There were 6 treatment-related deaths in the chemotherapy arm of this trial (3%). Altogether, 65% of the patients in the chemotherapy arm of the JBR10 trial completed at least three of their planned four courses of vinorelbine and cisplatin. There were 2 treatment-related deaths in the chemotherapy arm of the JBR10 trial (<1%). In the ANITA trial, the median percentage of planned chemotherapy doses was 76% for cisplatin and 56% for vinorelbine.⁸ The median number of chemotherapy cycles delivered was 4 for cisplatin (range 1–4) and 10 for vinorelbine (range 1–17). Deaths related to chemotherapy were observed in 1.7% of patients. In the CALGB 9633 trial, 85% of patients received their planned four courses of paclitaxel and carboplatin. There were no treatment-related deaths reported in the CALGB 9633 trial.

Efficacy of the Treatment for the Patients Participating in the Adjuvant Chemotherapy Trials

Four of the seven trials reported a reduction in the hazard ratio of death for the patients treated with adjuvant chemotherapy compared with those on observation (Table 76-2). Six of the seven trials used platinum-based chemotherapy similar to the regimen found to confer a survival benefit in the meta-analysis, whereas the sixth used tegafur and uracil (UFT). Three of these five trials using platinum-based chemotherapy (IALT, JBR10, and ANITA) and the trial using UFT showed a survival benefit for the treated patients compared with those on the observation arm.

It is difficult to compare the relative efficacy of the different chemotherapy regimens across the arms of the trials, and we await the future results of trials comparing different types of chemotherapy. The magnitude of the benefit for the European trials was small. There was no difference overall between the patients on the ALPI and BLT trials treated with resection plus chemotherapy versus those treated with surgery and observation. The reduction in the hazard ratio of death in the IALT trial (0.86; 95% confidence interval 0.76-0.98) was nearly identical to the reduction in the hazard ratio of death in the meta-analysis (0.87) compared with patients undergoing observation. This translated into a 4% absolute increase in the proportion of patients alive after 5 years (44.5% for those treated with adjuvant chemotherapy versus 40.4% for those on the observation arm), comparable with the 5% benefit seen in the meta-analysis.³ The reduction in hazard ratios of death for the chemotherapy-treated patients was remarkably similar in the other three trials. The hazard ratio of the patients treated with chemotherapy on the Japanese trial was 0.71, for the JBR10 trial it was 0.7 (p = 0.012), and for ANITA it was 0.79. This translated into a 3–15% absolute increase in the proportion of patients alive after 5 years for those treated with adjuvant chemotherapy compared with those on the observation arm.

Meta-Analysis

A meta-analysis combining individual patient data from five of the seven trials (ALPI, IALT, ANITA, BLT, and JBR10) has been published.⁴ These trials were chosen because they all used cisplatin-based chemotherapy as the adjuvant treatment and were the five largest adjuvant chemotherapy trials performed to date. In this pooled analysis, data were available on 4584 patients. The pooled analysis demonstrated a 4.2% increase in 5-year survival (hazard ratio 0.89, 95% confidence interval 0.82-0.96) favoring systemic chemotherapy. The benefits of adjuvant chemotherapy were demonstrated only for patients with stage II or III disease (hazard ratios of 0.83, 95% confidence interval 0.73-0.95, and 0.83, 95% confidence interval 0.73-0.95, respectively). There was no benefit for patients with stage IB disease (hazard ratio 0.93, 95% confidence interval 0.78-1.10) and a suggestion of a worse outcome with adjuvant chemotherapy for patients with stage IA disease (hazard ratio 1.41, 95% confidence interval 0.96-2.09), but there were only 377 patients (8%) with stage IA disease.⁴

Stage I NSCLC

The data from the adjuvant trials supports the use of chemotherapy for surgically resected stage II–IIIA NSCLC. However, the data for stage I NSCLC are less clear. The only trial that has shown a benefit for stage I NSCLC (stage IB only) is the UFT trial from Japan⁵ (see Table 76-2). However, this agent is presently not available in the United States. The CALGB 9633 study, which is the only study to have examined purely stage IB patients, demonstrated a survival benefit when initially reported in 2004 favoring chemotherapy. However, after additional follow-up of the study and reanalysis of the data in 2006, chemotherapy was no longer associated with a significant increase in survival (see Table 76-2). However, stage IB patients who received carboplatin-paclitaxel still had a prolonged progression-free survival compared with those randomized to the observation arm, and the survival is longer for those given adjuvant chemotherapy, but the difference reported in 2006 no longer achieves statistical significance, with the p values exceeding 0.05.⁹

DISCUSSION

These seven trials provide information about patients with NSCLC who benefit from adjuvant chemotherapy following surgical resection. Most information is reported by disease stage, but there is also some information on the benefit of adjuvant chemotherapy treatment by gender, age, histology, and type of chemotherapy, as well as on the impact of chest radiotherapy.

The most consistent benefit from adjuvant chemotherapy was observed in patients with resected stage II–III NSCLC. These studies (IALT, JBR10, and ANITA) used cisplatin-based chemotherapy as the adjuvant treatment. Two of the studies (JBR10 and ANITA) used cisplatin and vinorelbine. There are no data suggesting a benefit of carboplatin-based chemotherapy or from the use of a taxane in combination with cisplatin. Thus, based on these studies, the recommended chemotherapy in the adjuvant setting for patients with surgically resected stages II–IIIA NSCLC should be cisplatin and vinorelbine.

There are insufficient data to warrant the routine use of systemic chemotherapy for patients with surgically resected stage IA or B NSCLC. Analyses of the subset of patients with stage IB disease in the IALT, JBR10, and ANITA trials failed to demonstrate a benefit of chemotherapy. In addition, the CALGB 9633 trial, which was specifically conducted in stage IB patients, also failed to demonstrate a benefit of adjuvant chemotherapy. These findings do not rule out the possibility that chemotherapy is beneficial in stage I lung cancer. It is possible that the magnitude of benefit is quite small and that a very large dedicated clinical trial in stage IB NSCLC would be necessary to reveal such a benefit.

It remains a challenge to determine the appropriate postoperative treatment for patients with resected stage IIIA NSCLC. In addition to the studies using adjuvant chemotherapy (see Table 76-2), a prominent study looked at the effects of adding etoposide and cisplatin to chest radiation therapy for 488 patients with resected stage II and IIIA NSCLC.¹⁰ These patients did not benefit from the addition of chemotherapy. The clinical research to define the roles of chemotherapy and chest radiotherapy following resection of patients with stage III NSCLC has not yet provided a clear plan for their effective management.

Two studies show a therapeutic advantage and provide information about benefit by gender, age, and histology. The IALT and UFT studies show a benefit for both men and women, with no consistent impact by age. For patients on the IALT trial, the benefit was similar for both squamous cell and adenocarcinoma of the lung. The UFT trial enrolled only subjects with adenocarcinoma.

A future challenge will be to identify subsets of patients who derive the greatest benefit from adjuvant chemotherapy and those who do not benefit from adjuvant therapy. One example of this is K-ras mutations. These were specifically examined in the JBR10 trial. Patients with K-ras mutations did not derive any benefit from adjuvant chemotherapy. The effect of these mutations also should be examined in the other adjuvant clinical trials. Furthermore, tumor molecular characteristics may dictate the specific type of adjuvant therapy. For example, somatic mutations in the epidermal growth factor receptor (EGFR) have been associated with radiographic regressions to gefitinib and erlotinib in patients with advanced NSCLC.^11–13 Future studies may incorporate these agents in the adjuvant treatment of patients with surgically resected NSCLC bearing an EGFR mutation.

SUMMARY

Adjuvant chemotherapy is the standard of therapy for some patients with stages I–III breast and colon cancer. The therapeutic efficacy of adjuvant chemotherapy following surgical resection of early-stage NSCLC has been less clear. A meta-analysis was reported in 1995 of patients who underwent surgical resection for early-stage NSCLC and then were randomized to either observation or chemotherapy. This meta-analysis showed a 13% reduction in the hazard ratio of death, leading to a 5% absolute improvement in survival 5 years after the start of adjuvant cisplatin-based chemotherapy compared with observation only. Multiple prospective, randomized trials for patients with NSCLC were planned and undertaken to attempt to validate the observations of the meta-analysis. Seven trials with 150 or more patients with early-stage NSCLC (stages I–IIIA) on each arm have been reported in the last several years. Four of the seven trials show a survival advantage for the patients with early-stage NSCLC treated with adjuvant chemotherapy compared with those who underwent observation. The survival benefit in these four studies varies from a 4% to 15% survival advantage at 5 years after the start of chemotherapy. The hazard ratio of death for the patients treated with chemotherapy ranged from 0.61 to 0.86 compared with patients on observation. Thus the information available at the current time supports the administration of chemotherapy for patients with stages II and IIIB NSCLC. Further research will be needed to define the role of adjuvant chemotherapy for patients with stage I NSCLC.

EDITOR'S COMMENT

This is a new era in chemotherapy for lung cancer, where the paradigm for NSCLC stages IB through III now includes adjuvant chemotherapy. Recent reports show an improvement in five-year survival in patients receiving chemotherapy. Current trials are under way to assess the role of EGFR TK-inhibitor drugs or monoclonal antibodies to EGFR in addition to chemotherapy.

–MJK

REFERENCES