Small-Cell Lung Cancer
Thomas E. Stinchcombe
Mark A. Socinski
Lung cancer is the leading cause of cancer-related death among men and women in the United States, and it is estimated that in 2007 more people died from lung cancer than colorectal, breast, and prostate cancer combined.35 Small-cell lung cancer (SCLC) represents approximately 15% of the cases of lung cancer.26 The incidence of SCLC is decreasing within the United States, probably owing to a decrease in the prevalence of smoking.26 Among women, the incidence of SCLC is increasing, which reflects an increase in the prevalence of smoking among women approximately a generation ago.26 While tobacco use is the most common etiologic agent associated with SCLC, other such agents include exposure to asbestos, radon, uranium mining, and bis-chloromethyl ether.83 Cases of SCLC in lifelong nonsmokers have been reported, but the pathology should be carefully reviewed to confirm the diagnosis. SCLC is characterized by rapid growth, the high prevalence of early metastases to the mediastinal lymph nodes and distant sites, and high response rate to standard chemotherapy. Without treatment, the prognosis is very poor, and the median survival of patients who do not receive treatment is estimated to be 2 to 4 months.62
Staging
The proper staging of patients with SCLC determines the prognosis and treatment paradigm to be used. The most commonly used staging system is the two-stage system developed by Veterans Administration Lung Cancer (VALG) for patients with inoperable lung cancer.92 Limited-stage disease is defined disease that can be encompassed within a reasonable radiation field; extensive-stage disease is defined as disease that is greater than a reasonable radiation field. This simple staging system provides valuable prognostic information and remains the best predictor of response to therapy and survival.1,52,71
Approximately one-third of patients will present with limited-stage disease, and the standard treatment in appropriately selected patients is the combination of chemotherapy and radiation therapy, which has curative potential. Patients with extensive-stage disease have a high response rate to chemotherapy; however, the median survival is modest owing to the rapid development of chemotherapy resistance and subsequent progression of disease.
Despite the simplicity of this staging system, areas of controversy and ambiguity exist.46 Frequently limited-stage disease is considered disease confined to the one hemithorax—that is, the ipsilateral and contralateral mediastinal or ipsilateral supraclavicular lymph nodes. Patients with malignant ipsilateral pleural effusions are considered to have limited-stage disease according to the VALG definition; however, many cooperative groups and investigators consider these patients to have extensive-stage disease. The Radiation Therapy Oncology Group (RTOG) and the Eastern Cooperative Oncology Group (ECOG) have defined limited-stage disease as disease confined to one hemithorax and exclude patients with malignant effusion or contralateral hilar or supraclavicular lymphadenopathy.48 The National Comprehensive Cancer Network (NCCN) defines limited-stage as disease confined to the ipsilateral hemiothorax that can be safely encompassed within a tolerable radiation field.36 The European Organization for Research and Treatment of Cancer (EORTC) defines limited-stage disease as disease that involves <50% of the maximum transverse diameter of the thorax on a chest x-ray before starting chemotherapy.27 The tumor, nodal, and metastasis (TNM) staging system used by the American Joint Committee on Cancer (AJCC) is available for use in SCLC; however, it has not been widely adopted because most patients will present with advanced-stage disease at diagnosis, and surgical resection is rarely performed in this case.
The most frequent staging procedures are listed in Table 120-1. There can be significant variability on the staging procedures performed, because once a patient is determined to have extensive-stage disease, the utility of detecting additional sites of metastases may be limited and may not change the treatment paradigm. However, if the patient potentially has limited-stage disease, a meticulous workup for distant metastases should be performed. This includes a bone scan to evaluate for osseous metastases, computed tomography (CT) of the abdomen to evaluate for liver and/or adrenal metastases, and a magnetic resonance imaging (MRI) or CT scan with contrast of the brain to detect brain metastases. If a patient potentially has limited-stage disease and a pleural effusion, he or she should undergo a thoracentesis with cytologic evaluation to determine whether there is a malignant pleural effusion. Previously, bone marrow biopsy has been included as part of the staging workup; however, that is no longer routinely required. Bone marrow biopsy should be reserved for patients who have limited-stage disease and have a hematologic abnormality that is suspicious for bone marrow involvement or an unexplained neutropenia or thrombocytopenia.36 An elevated lactate dehydrogenase (LDH) has prognostic value, but it is not routinely included in the staging workup.1
Table 120-1 Standard Staging Test% | ||
|---|---|---|
|
Recently positron emission tomography (PET) scans with 18-fluoro-2-deoxy-D-glucose (FDG) have revealed superior sensitivity for detecting distant metastases in patients with non–small-cell lung cancer (NSCLC), and greater specificity for adrenal and bone lesions.15,43,89 A number of trials have investigated PET-scan staging in SCLC, but most of these trials have been performed at a single institution with a limited number of patients (≤120).7,8,10,37,65 Based on the preliminary evidence from these trials, PET-scan staging has been found to have greater sensitivity in the detection of intra and extrathoracic nodal involvement and distant metastases. A significant number of patients on these trials had their treatment plans changed owing to the detection of occult distant metastatic disease or changes in the radiation field to include intrathoracic lymph nodes with suspected malignant involvement. The current evidence-based medicine guidelines from the American College of Chest Physicians (ACCP) do not recommend the use of PET scanning in SCLC staging until additional studies have been performed that clarify the role of this modality.62 The current NCCN guidelines state that a PET scan can be used as part of the initial staging in addition to other recommended studies, and a bone scan is optional if a PET scan is performed.36
Surgical Treatment
The current standard treatment for limited-stage SCLC (LS-SCLC) is the combination of chemotherapy and radiation therapy. However, specific subsets of patients with LS-SCLC may benefit from surgical resection. In addition to these subsets, the SCLC histology can be an incidental finding at the time of surgery. An assessment of the role of surgical resection is complicated by the fact surgery is performed in only a small minority of patients, and these patients frequently receive additional chemo- and/or radiotherapy. There is also concern that the patients who are considered surgical candidates may have a biologically more indolent disease or would be considered good-prognosis candidates with chemoradiotherapy. Many of the case series or retrospective reviews of surgical therapy have included heterogeneous patient populations and were also performed before current staging procedures; thus a significant percentage of the patients who underwent surgery would have had metastatic disease detected with the current staging tests. Most of the interest in surgical resection in SCLC involves a subset of patients with LS-SCLC who present without evidence of nodal involvement or who are suspected of having a mixed histology of small- and non–small-cell lung cancer and have residual disease after chemoradiotherapy.
Several retrospective reviews have demonstrated long-term survival of patients with SCLC treated with surgery alone (Table 120-2).12,66,74,80 A randomized trial performed by the British Medical Research Council randomized operable patients (n = 144) to radiotherapy or surgical resection.22 The median survival was significantly longer on the radiotherapy arm in comparison to the surgical arm, 10 months versus 7 months, respectively (P = 0.04). This trial required bronchoscopic biopsy to confirm the diagnosis; thus there was a high prevalence of
centrally located tumors and, given the staging methods available at the time, undoubtedly many patients had distant disease at the time of surgical resection. On the surgical arm, only 48% of patients underwent surgical resection, raising doubts about the feasibility of surgical resection in LS-SCLC. The results of this trial discouraged surgical resection as the primary therapy for LS-SCLC.
centrally located tumors and, given the staging methods available at the time, undoubtedly many patients had distant disease at the time of surgical resection. On the surgical arm, only 48% of patients underwent surgical resection, raising doubts about the feasibility of surgical resection in LS-SCLC. The results of this trial discouraged surgical resection as the primary therapy for LS-SCLC.
Table 120-2 Role of Surgery in Small-Cell Lung Cancer | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The high prevalence of occult micrometastatic disease even in patients with early-stage disease and the high rate of distant failure led to interest in combining surgical resection and chemotherapy. Several retrospective reviews have investigated the survival of patients after a combination of surgical resection and chemotherapy and/or radiotherapy (Table 120-2).38,42,53,67,73 A retrospective review of patients who were treated with chemotherapy before and after surgical resection at the University of Toronto (n = 119) revealed a median survival of 111 weeks and a 5-year overall survival rate of 39%.70 The 5-year overall survival rate for patients with pathologic stage I (n = 35), stage II (n = 36), and stage III (n = 48) disease was 51%, 28%, and 19%, respectively. A separate retrospective analysis investigating the long-term survival (defined as ≥3 years) of 37 patients who underwent resection in comparison with 127 who did not undergo surgical resection revealed a long-term survival rate of 35% in patients who underwent resection and 6% in those who did not.30 All but one of the resected patients received chemotherapy, and all patients who did not undergo resection received chemoradiotherapy. The 5-year overall survival rate of patients with stage I disease (n = 13) was 64.2% and for stage II (n = 10), 42.5%. This survival data about patients with stage I and II disease have raised the possibility that surgery may have a role in the treatment of early-stage patients with SCLC.
Several other studies have investigated the role of preoperative or neoadjuvant chemotherapy.16,32,40,72,90 The use of preoperative chemotherapy may result in earlier treatment of distant micrometastatic disease and facilitate surgical resection by shrinking intrathoracic disease. A phase II trial at the University of Toronto (n = 72) prospectively evaluated the role of preoperative chemotherapy in patients with LS-SCLC.72 The overall response rate was 80%, and 57 patients (79%) were candidates for surgical resection, but only 38 patients underwent surgical resection. The median survival time for surgically resected patients was 91 weeks, and the projected 5-year survival rate was 36%. Patients with pathological stage I had a significantly longer survival time than patients with stage II or III disease. A second phase II trial (n = 40), which investigated preoperative chemotherapy, revealed an overall response rate of 87%, and 11 patients underwent thoracotomy and 8 underwent resection.57 Additional trials revealed a benefit to the combination of surgical resection after induction chemotherapy.6,44,69,72
In order to investigate the benefit of surgical resection after induction chemotherapy, the Lung Cancer Study Group (LCSG) performed a prospective randomized trial in which patients receiving five cycles of chemotherapy were randomized to radiotherapy to the chest and brain or surgical resection.40 Patients were required to have at least a partial response, have pure small-cell histologic features, and be appropriate candidates for surgical resection. Of the 328 patients enrolled, 217 patients achieved a response (66%) and 146 patients were randomized (44% of all patients). Of the patients randomized to surgical resection, 83% (n = 58) underwent surgical resection and 77% underwent complete resection (n = 54). The nodal status did not influence unresectability rates but the pretreatment T status influenced the unresectability rates; the unresectability rates for pretreatment T3, T2, and T1 were 40%, 16% and 6%, respectively. The median survival for all patients was 12 months and 16 months for patients who were randomized. There was no difference in survival between the treatment arms (P = 0.78). The results of this trial do not support the use of surgical resection after induction chemotherapy.
The ACCP guidelines state that no evidence-based medicine conclusions can be made regarding the role of surgical resection in early-stage SCLC.62 Our practice has been to consider surgical resection for patients who have T1 or T2 tumors and no radiologic evidence of mediastinal involvement or extrathoracic metastases.48 We will perform a mediastinoscopy for evaluation of the mediastinal lymph nodes to assess for the lack of mediastinal nodal involvement prior to pursuing surgical resection. SCLC has a greater propensity for mediastinal nodal metastases; thus data on the reliability of preoperative mediastinal staging in SCLC may not be as reliable as in NSCLC. In one small trial (n = 32) the false-negative rate of medianstinoscopy in SCLC was 16%.34 The presence of mediastinal lymph node involvement would preclude surgical resection. Patients who undergo surgical resection will then receive systemic chemotherapy and then be considered for thoracic radiation therapy (TRT) to reduce the rate of local relapse and consideration of prophylactic cranial radiation (PCI).48 The other clinical scenario when surgical resection is considered is when a patient has a mixed histology with SCLC and NSCLC and achieves a significant response but has residual disease, and the purpose of the surgical resection is to treat the residual NSCLC optimally.
The NCCN guidelines state that patients with T1 or T2 tumors should undergo mediastincoscopy or endoscopic evaluation of the mediastinum and, if negative, should undergo surgical resection, preferably a lobectomy with mediastinal nodal dissection or sampling.36 Postoperatively, all patients should receive chemotherapy, and patients with involvement of the mediastinum should receive concurrent chemotherapy and mediastinal radiation therapy.36
Limited-Stage SCLC
A recent analysis revealed that the percentage of patients who were classified as having limited-stage disease is approximately 40%.26 The vast majority of patients with limited-stage disease will be treated with a combination of chemotherapy and radiation.78 The 5-year survival rate from the United States National Cancer Data Base from 1985 to 2000 with this treatment approach is approximately 15%.24 In 2000, some 50% of the cases of LS-SCLC were in women, and 45% were in patients aged ≥70 years.24
The treatment paradigm of chemotherapy and radiation has been developed through multiple clinical trials over the last several decades. Treatment of LS-SCLC with chemotherapy alone results in poor local control rates, with intrathoracic failure occurring in 75% to 90% of patients.21 The addition of TRT significantly reduced the rate of intrathoracic failures to 30% to 60% but did not consistently result in an improvement in overall survival in individual clinical trials.21 In 1992, two separate meta-analyses were published establishing the role of TRT in the treatment of LS-SCLC.54,88 Pignon et al. evaluated 13 randomized trials of LS-SCLC, which included 2410 patients, and the relative risk of death in the chemoradiotherapy group compared with the chemotherapy-alone group was 0.86 (95%
confidence interval [CI], 0.78–0.94; P = 0.001).54 This corresponded to a 14% reduction in mortality and an absolute improvement in overall survival at 3 years of 5.4% ± 1.4%. No clear benefit was noted in the indirect comparison of early versus late TRT or sequential versus nonsequential strategies. Warde and Payne evaluated 11 randomized trials, and their analysis revealed an overall survival benefit for the addition of TRT to chemotherapy (odds ratio [OR] for 2-year survival = 1.53; 95% CI 1.30–1.76; P <0.001); the absolute difference in the 2-year survival rate was 5.4%.88 Data related to intrathoracic tumor control was available in 9 of the 11 trials, and when the addition of TRT was analyzed, it was found to have improved intrathoracic tumor control by 25.3% (OR for treatment effect on local control = 3.02; 95% CI 2.80–3.24; P <0.0001). The risk for treatment-related deaths was 1.2% (95% CI, -0.6%–3.0%). The consistent findings of these two meta-analyses established the role of TRT in the treatment of LS-SCLC and demonstrated that improvements in local control could result in improvements in overall survival. Cisplatin and etoposide (EP) and cyclophosphamide, doxorubicin, and vincristine (CAV) have demonstrated equal efficacy in extensive-stage disease; however, cisplatin and etoposide can be administrated concurrently with TRT, which has made it the preferred treatment paradigm in LS-SCLC.
confidence interval [CI], 0.78–0.94; P = 0.001).54 This corresponded to a 14% reduction in mortality and an absolute improvement in overall survival at 3 years of 5.4% ± 1.4%. No clear benefit was noted in the indirect comparison of early versus late TRT or sequential versus nonsequential strategies. Warde and Payne evaluated 11 randomized trials, and their analysis revealed an overall survival benefit for the addition of TRT to chemotherapy (odds ratio [OR] for 2-year survival = 1.53; 95% CI 1.30–1.76; P <0.001); the absolute difference in the 2-year survival rate was 5.4%.88 Data related to intrathoracic tumor control was available in 9 of the 11 trials, and when the addition of TRT was analyzed, it was found to have improved intrathoracic tumor control by 25.3% (OR for treatment effect on local control = 3.02; 95% CI 2.80–3.24; P <0.0001). The risk for treatment-related deaths was 1.2% (95% CI, -0.6%–3.0%). The consistent findings of these two meta-analyses established the role of TRT in the treatment of LS-SCLC and demonstrated that improvements in local control could result in improvements in overall survival. Cisplatin and etoposide (EP) and cyclophosphamide, doxorubicin, and vincristine (CAV) have demonstrated equal efficacy in extensive-stage disease; however, cisplatin and etoposide can be administrated concurrently with TRT, which has made it the preferred treatment paradigm in LS-SCLC.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
