Ninety percent of lung cancers are related to tobacco smoking, and the primary prevention of lung cancer by smoking cessation remains the primary goal of any physician involved in the field of diagnosis and treatment of lung cancer. However, given the rising trend in worldwide tobacco consumption and the fact that former smokers still have higher lung cancer risk than nonsmokers (in the United States, more than 50% of lung cancers occur in former smokers),¹ lung cancer will continue to globally represent a huge social health problem for a long period of time.

Lung cancer is the most common fatal malignancy among men and women in most countries throughout the world.²,³ Non-small cell lung cancer (NSCLC) represents more than 80% of all newly diagnosed cases of lung cancer. The reference treatment for early stage disease (from stage IA to IIB) is surgery; specific groups of patients with stage III disease may also benefit from pulmonary resection, usually in combination with other treatment modalities. Globally, the proportion of newly diagnosed NSCLC that can benefit from radical resection does not exceed 25% to 30% of the total number of cases.

The use of a systemic therapy in completely resected NSCLC is reasonably justified by follow-up studies after radical resection that have shown the predominance of distant failures over local recurrences and on some clinical and pathologic evidence of early microdissemination of the disease at the time of surgery.

Long-term survival in NSCLC following surgical resection is stage related, but even in stage IA, one third of patients will relapse and die of their disease within 5 years.⁴ Most of these relapses are distant metastases, whereas the risk of a local recurrence after complete resection is less than 10%. The central nervous system is the most common site of metastatic recurrence followed closely by bone, ipsilateral and contralateral lung, liver, and adrenals. More than 80% of recurrences occur within 2 years from the time of radical surgery.

The rate of recurrence for patients with stage II disease is higher than in stage I; more than 50% of resected stage II can be expected to relapse and, again, most recurrences are distant. The pattern of recurrence may differ by histology with more local recurrences seen for patients with squamous cell carcinoma and more distant metastases seen in patients with adenocarcinoma (Table 53.1).⁵,⁶,⁷,⁸

Positron emission tomography (PET) has been progressively implemented in the diagnostic workup for resectable NSCLC. Current experience indicates that metastatic disease will be found in 11% to 14% of the cases otherwise cleared for resection by conventional screening methods. In addition to distant metastases, almost all of these studies have demonstrated an increase in the rate of detection of unsuspected mediastinal and hilar nodal disease⁹,¹⁰,¹¹ (see Chapter 27).

Dissemination of cancer cells at much lower levels than those detected by any currently available imaging techniques, including PET scanning, seem to affect the prognosis of patients with clinical early stage NSCLC. Immunohistochemical and real-time polymerase chain reaction estimation of lymph node micrometastatic disease has been investigated in small retrospective studies evaluating the positivity for cytokeratins and carcinoembryonic antigen. Overall, the detection of positive findings in otherwise morphologically normal lymph nodes was almost invariably associated with an adverse outcome when compared with that of patients without occult micrometastatic disease.¹²,¹³,¹⁴,¹⁵

Quantification of free circulating DNA has also been proposed as a potential additional diagnostic tool for resected patients to detect persisting neoplastic disease.¹⁶

Although these studies are highly suggestive, additional confirmatory data are urgently needed to confirm the scientific hypothesis.

In addition to primary lung cancer prevention through smoking cessation campaigns, the detection of early stage disease through screening procedures represents the ultimate goal to defeat lung cancer. Unfortunately, this strategy remains investigational at this time and has not yet been proven to improve survival. Chest x-rays, analysis of cells contained in the sputum, and standard fiberbronchoscopy have shown limited effectiveness in early lung cancer detection. Newer tests
such as low-radiation, high-resolution helical computed tomography (CT) scan and molecular markers in sputum have been demonstrated to lead to earlier detection of lung cancers. However, their ability to increase survival rates of affected patients has not yet been fully accepted. Nonrandomized studies of high-resolution helical CT scan indicate an increase in the rate of detection of small resectable lung cancers, usually adenocarcinomas, and thus the frequency of lung surgery (see Chapters 15 and 16). However, as previously shown for chest x-ray screening, there may be neither a meaningful reduction in the number of advanced cancers being diagnosed nor a reduction in the number of individuals who die of lung cancer.¹⁷ This is because of the fact that CT screening can detect small nodules that may not be malignant or represent a group of slow-growing cancers that would not interfere with a patient’s life expectancy if they remained unnoticed.

Molecular markers are needed to identify tiny malignant nodules, which may be present among many nodules, often benign, that can be visualized by helical CT. Biomarkers are also needed for central lesions that cannot be identified by CT techniques, to indicate which patients should have bronchoscopic explorations to identify tiny intraepithelial central lesions. Automated, reliable, high-throughput sputum biomarker tests will ultimately replace cytomorphological techniques.

The Rationale for Adjuvant Treatments Following complete resection, tumor load, if any, is theoretically minimal. The relatively small number of residual neoplastic cells present in micrometastatic disease should contain few chemotherapyor radiation-resistant clones. The Gompertzian model of tumor growth and regression fits experimental and clinical data of most human solid cancers. If the assumption is correct, when the tumor is clinically undetectable, its growth rate should be at its largest and, although the numerical reduction induced by cytotoxic chemotherapy is small, the fractional cell kill from an effective dose of chemotherapy should be higher. In addition, pathological staging allows better prediction of prognosis and facilitates the comparison of treatment results between different trials.

How should the more appropriate treatment to be used in the adjuvant setting be selected? Realistically, no definitive rules have been established but, at least, the chosen treatment (or regimen) should be proven active in advanced disease, associated with good tolerability¹⁸; in the case of cytotoxic chemotherapy, it should be platinum based, initiated sufficiently early after radical surgery and administered for not less than three to four cycles.¹⁹

The Role of Adjuvant Radiotherapy For a long period of time, postoperative thoracic radiation therapy was the preferred adjuvant treatment. Results regarding its potential role have been reported from a large number of retrospective and prospective studies. Nine of these studies, collecting individual data from 2128 patients have been included in the Postoperative Radiation Therapy (PORT) metaanalysis and indicated postoperative radiotherapy as a treatment with significant detrimental effect on survival, especially in stages I and II.²⁰ These results have been further confirmed by a Cochrane systematic review and metaanalysis originally published in 2000 and substantially updated in 2004.²¹ The results of this recent update indicate a significant adverse effect of PORT on survival with a hazard ratio (HR) of 1.18 or 18% relative increase in the risk of death. This is equivalent to an absolute detriment of 6% at 2 years
(95% CI, 2% to 9%) reducing overall survival from 58% to 52%. Exploratory subgroup analyses suggest that this detrimental effect was most pronounced for patients with stage I/II, whereas for patients with stage III, N2, there was no clear evidence of an adverse effect.

Most of the studies included in these two metaanalyses incorporated patients treated with older technology (cobalt-60 and different dosimetry, and these outdated parameters may be partially responsible for the higher mortality rate observed in the radiotherapy group and attributed to an excess of intercurrent deaths. The use of newer technologies and improved dosimetry may prove to be effective as more recently suggested in a retrospective review.²² In addition, there were no sufficient data on the use of mediastinal lymph node dissection and surgical procedures, which differed from one study to the other and from one center to another. A more recent large scale analysis of PORT has been reported by Lally et al.²³ To investigate the association between survival and PORT in patients with resected NSCLC, 7465 patients coded as receiving PORT or observation with stage II or III NSCLC who underwent a lobectomy or pneumonectomy were selected within the Surveillance, Epidemiology, and End Results (SEER) database. Patients who survived less than 4 months were excluded. Median follow-up time was 3.5 years for patients still alive. Predictors for the use of PORT included age younger than 50 years, higher American Joint Committee on Cancer stage, T3 and T4 tumor stage, larger tumor size, advanced node stage, greater number of lymph nodes involved, and a ratio of lymph nodes involved to lymph nodes sampled approaching 1. On multivariate analysis, older age, T3 and T4 tumor stage, N2 node stage, male sex, fewer-sampled lymph nodes, and greater number of involved lymph nodes had a negative impact on survival. Overall, the use of PORT did not have a significant impact on survival. However, in subset analysis for patients with N2 nodal disease (HR = 0.855; 95% CI, 0.762 to 0.959; p = 0.0077), PORT was associated with a significant increase in survival. For patients with N0 (HR = 1.176; 95% CI, 1.005 to 1.376; p = 0.0435) and N1 (HR = 1.097; 95% CI, 1.015 to 1.186; p = 0.0196) nodal disease, PORT was associated with a significant decrease in survival. Hence, in this population-based SEER cohort, PORT use was associated with an increase in survival in patients with N2 nodal disease but not in patients with N1 and N0 nodal disease. A new large multi-institutional European phase III trial, Lung Adjuvant Radiotherapy Trial (Lung ART), compares threedimensional conformal PORT to no PORT, and will include patients who have proven N2 disease and a complete resection irrespective of whether adjuvant or neoadjuvant chemotherapy was used.²⁴ Seven hundred patients will be included to show a 10% difference in terms of 3-year disease-free survival (bilateral test, power = 80%, alpha error = 5%, from 30% to 40% at 3 years), which is the primary end point. With a longer follow-up period (median of 5 years), this sample size could also show a difference in survival rate of 9% at 5 years. The secondary end points will be overall survival, patterns of relapse, local failure, secondary cancers, and treatment-related toxicity. This project is being studied by the Intergroupe Francophone de Cancérologie Thoracique (IFCT), the European Organisation for Research and Treatment of Cancer (EORTC Radiation Oncology Group and Lung Group), and the Lung Adjuvant Radiotherapy Spanish Group.

Early Studies of Adjuvant Chemotherapy The history of adjuvant chemotherapy in completely resected NSCLC initiated in the early 1960s and 1970s with earlier trials testing the role of alkylating agents and nonspecific immunotherapies (mainly levamisole and Bacillus Calmette-Guérin [BCG]) that uniformly failed to demonstrate any survival benefit and, occasionally, a detrimental effect was observed.²⁵ All the drugs used in these studies had shown very limited or at all no activity in advanced NSCLC.

Subsequently, the potential role of cisplatin-based chemotherapy was extensively tested in all the stages of resectable NSCLC. All but one of these studies failed to show clinical benefit from adjuvant therapies.²⁶,²⁷,²⁸,²⁹,³⁰,³¹,³²

Common findings in both of these groups of studies include the overestimation of the potential benefit of adjuvant chemotherapy in the calculation of the sample size, in some trials, the unbalance in patients and treatment characteristics (for instance, incomplete mediastinal lymph node dissection), and, in most of these studies, the impossibility to reach the planned accrual. This probably reflects the negative attitude of thoracic surgeons toward adjuvant chemotherapy and the modern multidisciplinary approach to the patient with early NSCLC may be a way to overcome this problem.

In addition, most of the trials dose delivery including both dose and dose intensity were very often reported inadequate with an average of 50% of the patients receiving the full course of treatment.

In 1995, a metaanalysis overviewed eight cisplatin-based adjuvant chemotherapy studies, including all of the aforementioned studies, and demonstrated a 13% reduction of the risk of death, which was close to the borderline of statistical significance (p = 0.08). Similarly, there was a 6% reduction in the risk of death in patients treated with postoperative radiotherapy and cisplatin-based chemotherapy compared with patients who received only postoperative radiotherapy (p = 0.46). Conversely, adjuvant chemotherapy with long-term alkylating agents was significantly detrimental.³³

These findings failed to impact on clinical practice not because the absolute gain was too small, but because such an estimate was still imprecise, ranging from 1% detriment to a 10% benefit. In addition, the heterogeneity of surgical procedures, specific chemotherapy regimens, difference in the staging modalities, and absence of a single large prospective trial truly demonstrating increased survival as a result of the use of chemotherapy limited the applicability of the results of this metaanalysis. However, these data strongly supported additional prospective testing of
adjuvant chemotherapy, and new trials using state-of-theart chemotherapy were initiated.

Recent Platinum-Based Adjuvant Chemotherapy Trials The aforementioned metaanalysis generated enthusiasm to prompt the planning of several prospective randomized phase III studies, all platinum based (positive/negative thoracic radiotherapy), evaluating the role of modern platinum-based regimens in all resectable stages of NSCLC.

The North American Intergroup trial evaluated the efficacy of four cycles of postoperative cisplatin/etoposide plus concomitant thoracic radiotherapy (total dose of 50 Gy) in comparison with PORT alone in stage II and IIIA NSCLC. A total of 463 patients were included with no significant difference between the two arms in terms of median time to progression. The relative likelihood of survival among patients assigned to receive chemotherapy plus radiotherapy, as compared with those assigned to receive radiotherapy alone, was 0.93 (95% CI, 0.74 to 1.18).³⁴ Toxicity of radiation caused by the concomitant administration of cytotoxic agents may explain the lack of efficacy (more striking in stage II). Biological correlative studies evaluating differential expression of p53 and K-ras did not show any relationship with outcome.³⁵ This study was unique in its routine use of PORT. All other studies primarily tested the efficacy of adjuvant chemotherapy versus observation alone in completely resected NSCLC, with the optional administration of sequential PORT in some of them according to an investigator choice (Table 53.2).

From 1994 to 1999, a joint effort from Adjuvant Lung Project Italy (ALPI) and EORTC enrolled 1209 patients with completely resected stage I, II, or IIIA NSCLC.³⁶ Patients were randomly assigned to receive chemotherapy with mitomycin, vindesine, and cisplatin (MVP) for three cycles or observation. A total of 69% completed three cycles of MVP, with half of those patients requiring dose reduction. Radiotherapy was given according to the policy of individual centers: 43% of patients received PORT. No significant difference in overall survival was seen with an HR of death of 0.96. Median overall survival was 55 months in the chemotherapy arm and 48 months in the surgery arm. Subset analysis by stage showed that HR was 0.80 (95% CI, 0.60 to 1.06) for stage II versus 0.97 (0.71 to 1.33) and 1.06 (0.82 to 1.38) for stages I and III, respectively. It is remarkable that in the subgroup of patients with stage II NSCLC, although the hazard ratio was not statistically significant, a 10% survival advantage at 5 years for chemotherapy-treated patients was reported.

Although the incidence of grades 3 and 4 hematological and nonhematological toxicities related to chemotherapy did not differ quantitatively and qualitatively from those commonly reported in advanced NSCLC, the marginal reduction in survival observed in the MVP arm in the first year after randomization could potentially reflect a toxicity effect. This is also indirectly confirmed by the lower percentage of patients in the MVP arm who completed subsequent thoracic radiotherapy (65% vs. 81% in the control arm).

No statistically significant association between p53 or Ki67 expression and stage or histology was found. An analysis of K-ras mutation status and survival was performed in adenocarcinomas and large cell carcinomas: mutations were found in 22% of the 117 considered samples with no relationship to survival.

The International Adjuvant Lung Cancer Trial (IALT) collaborative group was the first large study to show a significant benefit in favor of adjuvant chemotherapy. A total of 1867 completely resected NSCLC patients were randomized to chemotherapy (a doublet of cisplatin plus vindesine, vinblastine, vinorelbine, or etoposide) or observation.³⁷ All stages were represented, with approximately 10% having stage IA disease, 27% stage IB, 24% stage II, and 39% stage III. In the chemotherapy arm, 74% of patients received at least 240 mg/m² of cisplatin, 27% of patients received PORT. Toxicity of grades 3 and 4 was experienced by 23% of patients (0.8% of toxic deaths). Survival was significantly longer in the chemotherapy arm with an HR of 0.86 (95% CI, 0.76 to 0.98; p <0.03): 5-year survival rates were 44.5% and 40.4%, and median survival was 50.8 and 44.4 months in the chemotherapy group and in the control group, respectively, whereas median diseasefree survival was 40.2 and 30.5 months.

Some methodological aspects regarding the final analysis of the IALT study raise concerns: in clinical trials in which a long follow-up is required, the difference between treatments may depend on the follow-up time. That implies that in the early phase of the study, there is the potential for a biased estimate of the treatment effect. In the IALT study, the accrual was prematurely interrupted when less than 60% of expected patients had been enrolled. The follow up continued until about 65% of expected events had been observed. If the study had continued until the planned number of events was reached, the conditional power to detect, under the null hypothesis, a statistically significant difference would have been less than 50%. Moreover, the adoption of a Bayesian approach that may be more appropriate for interpreting results, when early analysis shows a potential positive treatment effect, would have suggested to prolong follow-up.³⁸ Bayesian analysis to determine if the IALT results were convincing enough to change clinical practice were presented at the American Society of Clinical Oncology (ASCO) 2004 and the preponderance of evidence supported at least a 3% survival advantage for adjuvant therapy.³⁹ Of interest, however, is that the results of the IALT study, initially reported with a 4.7-year follow-up (ASCO 2003), were recently updated (ASCO 2008) with 3 additional years of follow-up.⁴⁰ Median follow-up was 7.5 years at the cutoff date of September 1, 2005. The survival status was known for 1807 patients. Results showed a beneficial effect of adjuvant chemotherapy on overall survival (HR = 0.91; 95% CI, 0.81 to 1.02; p = 0.10) and on disease-free survival (HR = 0.88; 95% CI, 0.78 to 0.98; p = 0.02). However, there was a significant difference between the results of overall survival before and after 5 years (HR = 0.86; CI, 0.76 to 0.97; p = 0.01 vs. HR = 1.45; CI, 1.02 to 2.07; p = 0.04); p value for interaction was 0.006. Disease-free survival benefit was also different according to the follow-up duration (p value for interaction: 0.04; global, first 5 years, HR = 0.85, p = 0.006; after 5 years, HR = 1.33, p = 0.16). The analysis of non-lung cancer deaths for the whole period showed an HR of 1.34 (CI, 0.99 to 1.81; p = 0.06). These results confirmed the efficacy of chemotherapy for the first 5 years after surgery. The difference in results between less than and more than 5 years of follow-up may suggest possible late adjuvant chemotherapy-related overmortality. This potential effect underscores the need for the long-term follow-up of adjuvant lung cancer trials in order to evaluate results in terms of treatment benefits and long-term hazards.