CHAPTER 19 Lung Cancer
Multimodal Therapy
Though long regarded as the best chance for cure, surgery alone has shortcomings in the management of non–small cell lung cancer (NSCLC). Even in the most favorable of circumstances (stage I), failure of therapy is anticipated in 20% to 40% of patients.1–3 Moreover, because most patients do not have localized disease at the time of the initial diagnosis, surgical resection is offered to only a minority of all patients with NSCLC.
Predicting which patients will not be cured by resection permits identification of appropriate candidates for multimodality therapy. To this end, locoregional lymph node involvement is a powerful predictor of cancer recurrence and should be thoroughly investigated during treatment planning.4 Tumor histology and size also affect cancer-related and overall survival of resected patients.5–7 Moreover, numerous genetic markers are being discovered that may aid outcome prognostication, and when combined with current imaging modalities (e.g., positron emission tomography), outcome prediction may be further refined.8–10 Recent reports have proposed systems using histologic and molecular criteria to stratify predicted survival of resected patients.11–13
In developing strategies for combined therapies, patterns of failure must be considered. Locoregional recurrence after resection of stage I cancers is rare, but it is more common when regional lymph nodes are involved. However, because patients seldom die from local recurrence, the impact of improving local control with adjuvant radiotherapy may not translate to a survival advantage. Even in the presence of local lymph node involvement, resected patients with stage II cancers are still twice as likely to fail systemically as locally,14 suggesting that adjuvant systemic therapy (chemotherapy) might be more useful than adjuvant local therapy (radiotherapy) and, if effective, may ultimately result in a survival advantage.
ADJUVANT THERAPY
Despite enthusiasm for multimodality treatment of NSCLC, and after more than 35 years of promising clinical trials, it has only recently become apparent that there is a clear and consistent benefit of adjuvant therapy for resected lung cancer. The delay may have been partly because chemotherapy and radiation protocols changed frequently, resulting in large numbers of studies that were difficult to compare. Few studies have been designed to critically examine treatment effects on specific NSCLC stages; most studies tend to group stages I through III in their treatment arms. This shortcoming has been compounded by a lack of accurate pathologic staging in most trials. Moreover, promising results generated at high-volume academic centers may not be easily reproduced at less specialized smaller institutions.15,16
Adjuvant Chemotherapy
Early randomized trials of adjuvant chemotherapy focused on the use of nitrogen mustard,17,18 cyclophosphamide,17,19 and a combination of lomustine and hydroxyurea.20 Cumulatively, these trials demonstrated that treated patients had more postoperative complications without any survival benefit than did untreated patients. In fact, patients treated with postoperative cyclophosphamide experienced a poorer long-term survival than did untreated patients.19 This was independently confirmed by meta-analysis.20 Findings from these early studies have been largely invalidated because histologic type and pathologic stage were not considered during trial design. Moreover, chemotherapeutic agents used in these early trials have since been demonstrated to be ineffective, if not detrimental, for patients with NSCLC.
As several studies began documenting the efficacy of platinum-based agents for advanced (stage IV) NSCLC, platinum-based adjuvant in combination chemotherapy regimens became the preferred choice for resected patients. Initial enthusiasm was generated when the Lung Cancer Study Group trial 772,21 a randomized study, suggested a disease-free and overall survival benefit in the treatment arm. Although the survival advantage was not considered statistically significant, sufficient optimism was generated to continue investigations of these types of chemotherapeutic protocols. Early subsequent trials, however, proved to be far less promising, although sufficient flaws were identified in the design or conduct of each to rationalize their failure.
A randomized study of cyclophosphamide, Adriamycin, and cisplatin adjuvant therapy for resected stage I to II NSCLC22 did not demonstrate treatment efficacy. This trial suffered from the inability to deliver the prescribed chemotherapeutic regimen (four cycles) over the prescribed time. Fewer than 30% of patients in the treatment arm actually received chemotherapy as intended. Other randomized studies have been similarly plagued by incomplete delivery of prescribed postoperative therapies,23,24 which partly contributes to the negative findings of these trials. Additionally, the dose of cisplatin (40 to 60 mg/m2/cycle) was lower than that currently recommended (80 to 120 mg/m2/cycle).25
Although all of these early platinum trials were considered statistically negative, subtle survival differences were noted between control and treated patients in essentially every study. Not surprisingly, a 1995 meta-analysis of several randomized adjuvant platinum-based trials demonstrated a 13% reduction in the risk of death (P = .08) for treated patients.26 This rekindled interest in this adjuvant strategy and spawned multiple modern trials. Platinum dosage was adjusted so that patients were scheduled to receive 300 to 400 mg/m2 (total dose). Adjuvant radiation therapy was left to the discretion of participating centers. Two studies demonstrated trends favoring survival benefit in treated patients, although neither was considered to be a positive trial.27,28 The International Adjuvant Lung Cancer Trial,29 however, demonstrated a clear survival benefit of adjuvant platinum-based chemotherapy. This study included resected stage I to III patients and found a survival advantage, 5% percent benefit after 5 years (P = .003), favoring the adjuvant chemotherapy arm versus surgery alone. This survival advantage is similar in magnitude to that achieved by adjuvant chemotherapy for resected breast and colon cancers.30,31
The Lung Adjuvant Cisplatin Evaluation (LACE) pooled data from the five largest randomized trials completed since 2004 and reanalyzed this collective.32 This study amassed information from more than 4500 patients who were previously randomized to treatment and concluded that postoperative cisplatin-based chemotherapy significantly improves survival in patients with NSCLC. Because the index trials feeding LACE were heterogeneous with regard to stage (I to III), subgroup analysis of the pooled data allowed some insight into the differential effect of adjuvant therapy on specific stages of resected cancer. From LACE, there is little doubt that resected stage II and III patients will enjoy a meaningful survival advantage from adjuvant chemotherapy—at least 5% for overall survival and 6% for disease-free survival.32 Patients with resected stage IB disease, however, have a much less obvious benefit, although the trend is toward some advantage. The confidence intervals of the hazard ratio are wide and quite similar to those of the final analysis of the Cancer and Leukemia Group B (CALGB) 9633, a trial of adjuvant carboplatin and paclitaxel for resected stage IB patients only.33 An important finding of CALGB 9633 was that patients with primary tumors larger than 4 cm do, in fact, benefit from adjuvant chemotherapy.33
Adjuvant Radiotherapy
Currently, there is no justification to include adjuvant radiotherapy in the management of resected stages I and II NSCLC. Three randomized trials,34–36 as well as meta-analyses of trials (postoperative radiotherapy [PORT] studies),37,38 document a deleterious effect (a survival disadvantage) of adjuvant radiotherapy for early-stage NSCLC. Reasons for this include poor radiotherapy planning,36 larger fraction size,34 greater total radiation dosage,34–36 outdated equipment, and poor quality control.39
Although there was optimism that hilar lymph node (N1)-positive patients (stage II) would provide a good substrate for adjuvant radiotherapy, this appears not to be the case. From subgroup analysis of several randomized trials,37,38,40,41 radiotherapy has no significant effect on either local control or overall survival for these patients.
Finally, there are insufficient randomized data to defend the use of adjuvant radiotherapy for resected stage III (N2/N3) disease. Even though improved local control has been reported,40–42 it has not translated into a meaningful survival benefit. The strongest argument to support adjuvant radiotherapy for resected stage III NSCLC comes from a nonrandomized, single-institution, retrospective study.17,43 These investigators found that postoperative radiotherapy was the strongest independent predictor of survival for resected N2-positive disease. This study, however, has been widely criticized,39 and PORT meta-analyses consistently do not demonstrate clear efficacy for adjuvant radiotherapy for stage III disease.37,38 Yet, controversy still remains over radiotherapy used in this setting, as one recent trial of adjuvant chemotherapy demonstrated the best survival when patients received thoracic radiation after their chemotherapy for resected stage III disease.44
INDUCTION THERAPY
Use of chemotherapy in the induction setting has several potential advantages. First, there is earlier treatment of micrometastatic systemic disease. Because distant recurrence is the most common mode of failure, early eradication of systemic disease might translate into a survival advantage, and chemotherapy is most efficacious when the tumor burden is small.45 Also, drug delivery and cytotoxic effects may be enhanced as a result of preservation of tumor blood supply. Perhaps most importantly, however, a higher percentage of patients will probably receive their intended dose of chemotherapy when drugs are delivered preoperatively.46
Induction radiotherapy in combination with chemotherapy has been and continues to be investigated vigorously. Two large phase III trials comparing only induction radiotherapy (40 to 50 Gy) were reported several decades ago.47,48 Both demonstrated inferior survival rates for radiated patients. Since then, limited data have emerged to suggest a role for radiotherapy alone as an induction modality. Consequently, discussions of induction radiotherapy are primarily in the context of chemoradiotherapy protocols.
Induction therapies are not without concerns. Subsequent resections are generally more technically challenging and may result in lung-sparing operations becoming less feasible. Morbidity and mortality rates for surgery after induction therapy may be greater,27,49 and for an early-stage cancer, given the expected relatively good survival, the risks may well exceed any benefits. Despite these concerns, several groups are pursuing rigorous induction strategies and demonstrating reasonable safety profiles.50,51
Induction Chemotherapy for Early-Stage Non–Small Cell Lung Cancer
Observations from other solid tumor treatment trials suggest that chemotherapy response rates improve as similar drug regimens are applied to earlier stages of disease.52 The feasibility of induction chemotherapy for early-stage lung cancer was demonstrated in 2000 by the Bimodality Lung Oncology Team study.46 This phase II trial examined the response, toxicity, resectability rate, surgical morbidity, and intermediate survival for patients treated preoperatively with carboplatin and paclitaxel. Two preoperative cycles were administered and three postoperative cycles planned for patients undergoing complete resections. After induction therapy, 56% of patients had a major objective response; 86% of patients underwent complete resection. Impressively, 96% of patients received the intended preoperative chemotherapy, and 46% received the planned postoperative courses. Treated patients appeared to have standard postoperative recoveries.
A more recent randomized trial suggests a benefit of induction chemotherapy for stages I and II NSCLC.27 The induction regimen consisted of two cycles of mitomycin, ifosfamide, and cisplatin. Two additional cycles were given postoperatively for responding patients. The patient population was heterogeneous, with some stage IIIA patients included in the study. Nonetheless, despite a slightly higher mortality rate for chemotherapy-treated patients, disease-free survival was longer than for untreated patients with stage I or II cancers.
Induction Therapy for Stage IIIA/B Non–Small Cell Lung Cancer
Induction Chemotherapy
In 1982, Pearson and colleagues reported on the dismal outcome of surgery alone for stage IIIA (N2 lymph node–positive) patients.53 They noted a 9% 5-year survival rate for gross mediastinal lymph node involvement and a 24% survival rate for patients with microscopically positive nodes. Even for occult, single-station N2 disease, the best reported survival rate is less than 30% after 5 years.54 Local recurrence seldom contributed to mortality, because the vast majority of these patients die of metastatic disease. Consequently, induction chemotherapy as the systemic component of multimodality treatment for patients with locally advanced NSCLC has been investigated.
By the mid 1990s, the recognized standard of care for resectable stage IIIA (N2-positive) NSCLC was changed by the near-simultaneous reports of two small randomized trials designed to study the efficacy of induction chemotherapy followed by resection. Until this time, most had considered combined definitive chemoradiotherapy as the standard of care. Collectively, these two reports represented data acquired from only 120 patients. Rosell and associates55 compared surgery alone with three cycles of induction chemotherapy (platinum based) followed by surgery. All patients in the study were scheduled to receive postoperative thoracic irradiation. A second, more chemotherapy-intensive trial consisted of three cycles of a cisplatin regimen given as induction with three additional cycles planned postoperatively for responding patients.56 In this trial, radiotherapy was reserved for unresectable or incompletely resected patients. Entry into either trial required pathologic confirmation of mediastinal lymph node involvement (N2 disease) by mediastinoscopy, or T3 disease.
Although neither study was without shortcomings, surprisingly both trials were stopped early because interim analyses demonstrated a significant advantage favoring patients given induction chemotherapy. Mature, actual (not actuarial) survival updates have been published, with an average of 7 years of follow-up.57,58 These reports document a durable and statistically significant survival benefit of induction chemotherapy with surgery versus surgery alone, and they translate into a 20% better 5-year survival rate.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
