During the past 3 decades, progress has inched forward in the management of limited small-cell carcinoma of the lung. Prior to the 1970s, patients were managed largely with surgery alone or radiation therapy alone. In 1969, the British Medical Research Council reported 5-year survival rates of only 1% with surgery and 4% with radiotherapy.
In the early 1970s, the ability of small-cell carcinoma to disseminate early prompted clinical trials focusing on systemic management with chemotherapy. Initial results with chemotherapy, yielding a four- to fivefold increase in median survival, generated great enthusiasm and the hope that outcomes similar to those achieved with other chemosensitive tumors would be forthcoming. Now, almost 30 years later, high rates of local and distant relapse, including central nervous system relapse, continue to thwart efforts to achieve this elusive goal.
Nevertheless, real progress has been made, and the use of combination chemotherapy, concurrent thoracic radiotherapy, and prophylactic cranial irradiation have led to 5-year survival rates as high as 26%. New chemotherapeutic agents with significant activity against small-cell carcinoma are emerging, and refinements in radiotherapeutic technique are producing not only improvements in local control, but survival benefits as well. This article will review the current approach to limited small-cell carcinoma of the lung, and will explore unanswered questions and future directions.
The overwhelming tendency of small-cell carcinoma of the lung is to disseminate early. This has taught us that even when a rigorous search fails to find distant metastases, the best approach is to treat limited small-cell carcinoma of the lung as a systemic disease, with chemotherapy as the cornerstone of treatment. As a result, there is less emphasis on the TNM classification, which is more appropriate when surgery is being considered. Instead, a simplified staging system of limited disease vs extensive disease is used.
The limited-disease category includes patients whose disease may be encompassed within a radiation portal. Ambiguities about pleural effusion and nodal stations cause some confusion in categorizing small-cell lung cancer as limited or extensive. The presence of an ipsilateral pleural effusion frequently, though not uniformly, excludes patients from limited-disease protocols, although technically this is considered limited disease. Some investigators include within the limited-disease category patients with minimal pleural effusions not felt to be easily accessible for cytologic diagnosis; this includes those with blunting of the costophrenic angle on chest radiographs, as well as those with effusions seen only on chest computed tomography (CT). Many investigators,[7-9] however, exclude all patients with demonstrated pleural effusions by any study, including chest CT.
Similar variability in the definition of limited disease occurs with reference to the extent of lymphadenopathy. Some studies include patients with bilateral supraclavicular adenopathy in the limited-disease category; others[4,6] include only those with ipsilateral adenopathy. Likewise, some investigators include patients with contralateral hilar adenopathy in the limited-disease category; others exclude them. These variations reflect a change in approach from a time when all nodal stations were targeted by radiotherapy ports to the more recent policy of including only nodes with obvious involvement radiographically.
In addition to the variability in the definition of limited-stage disease, current imaging techniques detect more extrathoracic disease than did planar images. In many of the older studies, thoracic imaging consisted only of chest radiographs.[7,11,12] Current staging minimally includes chest CT extending through the liver and adrenals (Table 1). Because the brain is a common site of metastatic disease in small-cell lung cancer, imaging of the brain by CT or magnetic resonance imaging (MRI) remains sensible. A radionuclide bone scan is also frequently performed. However, bone marrow aspiration and biopsyonce a routine staging procedureis no longer required. Bone marrow involvement as the sole manifestation of extensive disease is quite rare, occurring in only 1.7% of patients. Thus, this invasive procedure, and the associated discomfort, may be omitted.
At present, positron-emission tomography (PET) has no defined role in the routine staging evaluation of small-cell carcinoma patients. However, use of this highly sensitive imaging modality is increasingly being explored in a variety of oncologic settings. It has the potential to become a very useful staging procedure for small-cell carcinoma patients and may prove beneficial in terms of monitoring their response to therapy. The precise role of PET scanning in small-cell carcinoma will emerge in the next decade.
As imaging modalities continue to improve, the ability to detect disease outside of what might be encompassed within a radiation portal will continue to increase. Patients who would previously have been classified as having limited-stage disease will be found to have extensive-stage disease. This stage migration will appear to improve the outcome of both limited- and extensive-disease patients, another example of the Will Rogers phenomenon.
In the 1970s, chemotherapy regimens consisted of single-agent alkylators and then combinations based on alkylating agents (mainly cyclophosphamide(Drug information on cyclophosphamide) [Cytoxan, Neosar]). One of the most commonly prescribed regimens throughout the 1980s was CAV (cyclophosphamide, doxorubicin(Drug information on doxorubicin) [Adriamycin], and vincristine [Oncovin]). Doxorubicin was considered highly effectivethe paclitaxel(Drug information on paclitaxel) (Taxol) of the 1970s. This regimen produced excellent response rates, but the majority of patients relapsed both locally and systemically. Also, the inclusion of doxorubicin created havoc in regimens combining chemotherapy with radiation therapy, because doxorubicin potentiated radiation-induced toxicity and caused radiation recall.
Cisplatin and Etoposide(Drug information on etoposide)
In the late 1970s, a regimen consisting of cisplatin(Drug information on cisplatin) (Platinol) and etoposide was developed.[16,17] Preclinical studies suggested a marked synergy with this combination, whereas single-agent therapy with cisplatin had produced a response rate of only 10%. The combination was studied as salvage therapy for patients with recurrent or refractory small-cell carcinoma and was associated with response rates as high as 52%. Cisplatin/etoposide was investigated as first-line therapy, and was found to be highly active and amenable to combination with concurrent thoracic irradiation. The regimen proved to be equivalent or superior to all previous combinations and in the 1980s became the treatment of choice for limited small-cell carcinoma.
A prospective, randomized trial of standard-dose cisplatin/etoposide (cisplatin, 80 mg/m² IV on day 1, and etoposide, 80 mg/m² IV on days 1 to 3, repeated every 3 weeks) compared to high-dose cisplatin/etoposide (cisplatin, 27 mg/m² IV on days 1 to 5, and etoposide, 80 mg/m² IV on days 1 to 5, repeated every 3 weeks) showed no improvement in efficacy and a substantial increase in toxicity in the high-dose arm. Even standard-dose cisplatin regimens, however, produce significant toxicity, most notably nausea, vomiting, nephrotoxicity, and neuropathy.
Carboplatin(Drug information on carboplatin) (Paraplatin) may have a more favorable toxicity profile, and has been used instead of cisplatin. The carboplatin/etoposide combination has also demonstrated excellent activity in small-cell carcinoma, and in a prospective randomized phase III trial conducted by the Hellenic Cooperative Oncology Group, carboplatin/etoposide was associated with equal efficacy and less toxicity than the cisplatin/etoposide combination.
Because of the equivalent efficacy and favorable toxicity profile of carboplatin, many clinicians prefer carboplatin/etoposide over cisplatin/etoposide. The combination of a platinum compound with etoposide remains standard therapy for small-cell carcinoma. Over the past few years, however, several new cytotoxic agents with substantial activity in this disease have been developed.
Paclitaxel was introduced in 1993, and phase II studies demonstrated considerable single-agent activity in previously untreated and treated small-cell lung cancer patients (Table 2).[23-25] Greco and Hainsworth added paclitaxel as a 1-hour infusion to a commonly used combination of carboplatin/etoposide. Hainsworth et al started with modest doses of paclitaxel (135 mg/m²) and carboplatin (area under the concentration-time curve [AUC in mg/mL · min] = 5), but because the myelosuppression that developed with this regimen was manageable, the investigators subsequently treated a larger number of patients with increased doses of paclitaxel (200 mg/m²) and carboplatin (AUC = 6). This study included previously untreated small-cell carcinoma patients with limited or extensive disease. Limited-disease patients received thoracic irradiation at 1.8 Gy/d to a total dose of 45 Gy over 5 weeks, beginning concurrently with cycle 3 of the chemotherapy. While response and toxicity data were promising, this three-drug combination is costly, and whether it will prove superior to the combination of carboplatin/etoposide chemotherapy awaits the results of a prospective, randomized trial.
Paclitaxel has also been combined with the cisplatin/etoposide regimen. This combination has produced a response rate of 94% in extensive-disease patients. Recently, a multi-institutional phase I/II study of this regimen administered with concurrent thoracic irradiation to limited-disease small- cell carcinoma patients was published. In this trial, four 21-day cycles of chemotherapy were administered, concurrently with thoracic irradiation given at a total dose of 45 Gy over 5 weeks, beginning on day 1 of cycle 1. Cisplatin, 60 mg/m², was given on day 2 of all cycles. Etoposide was given at a lower dose, 60 mg/m²/d, on days 1 to 3 of cycles 1 and 2 (with concurrent radiation), and at a higher dose, 80 mg/m²/d, on days 1 to 3, during cycles 3 and 4. Granulocyte-colony stimulating factor (G-CSF [Neupogen]) was added during cycles 3 and 4.
During the phase I portion of the trial, the paclitaxel dose during cycles 1 and 2 was escalated to determine the maximum tolerated dose with concurrent radiation. This was found to be 135 mg/m², administered intravenously over 3 hours on day 1; grade 4 neutropenia was the dose-limiting toxicity. During cycles 3 and 4, paclitaxel was given at 170 mg/m². The overall response rate for this regimen was 96%, with 39% complete responses. Again, the question of whether this active, but costly, three-drug regimen is superior to standard cisplatin/etoposide in terms of survival cannot be answered without a prospective, randomized trial.
Topoisomerase I Inhibitors
The topoisomerase I inhibitors, topotecan(Drug information on topotecan) (Hycamtin) and irinotecan(Drug information on irinotecan) (Camptosar), have also been shown to have significant activity against small-cell carcinoma. Topotecan (Table 3) has been studied in the salvage setting, where it yielded response rates of 14% to 38% in patients with sensitive diseaseie, those who responded to first-line chemotherapy and subsequently relapsed more than 3 months after their chemotherapy was discontinued.[30-33] Topotecan was noted to be considerably less effective in patients with refractory small-cell carcinoma, where response rates ranged from 2% to 11%.[30-32,34] In previously untreated extensive-disease patients, a response rate of 39% has been reported.
An ongoing phase I trial is being conducted to determine the maximum tolerated systemic exposure of topotecan when combined with carboplatin/etoposide in extensive-disease patients. Preliminary results show an 81% response rate.
The combination of topotecan and paclitaxel has been shown to be active by several investigators (Table 4).[37-39] A study in limited-disease patients is currently being conducted by the Cancer and Leukemia Group B (CALGB). In this protocol, patients initially undergo two cycles of induction chemotherapy with the new combination of topotecan/paclitaxel with G-CSF support. They subsequently receive the standard regimen of carboplatin/etoposide for three cycles. Thoracic irradiation is given starting with the first cycle of carboplatin/etoposide: A 60-Gy total dose is being administered to the first 10 patients; then, if well tolerated, a 70-Gy total dose. Prophylactic cranial irradiation is required for those who achieve complete or very good partial remissions.
Irinotecan has also shown substantial activity in previously treated small-cell carcinoma patients (Table 5).[40-42] A phase II study, including both limited-disease and extensive-disease patients, evaluated the combination of irinotecan, 60 mg/m² on days 1, 8, and 15, with cisplatin, 60 mg/m² on day 1, every 28 days. Patients with limited disease received four cycles of chemotherapy followed by thoracic irradiation to a dose of 50 Gy. The response rate in limited-disease patients was 83%, with 30% complete remissions; median survival was 14.3 months. The major toxicities were myelosuppression and diarrhea.
This chemotherapy regimen has also been explored with concurrent thoracic irradiation in limited-disease small-cell lung cancer patients. In this setting, the dose-limiting toxicity was fatigue, and the recommended regimen for future study is irinotecan, 40 mg/m² on days 1, 8, and 15, with cisplatin, 60 mg/m² on day 1, every 28 days. Whether this combination will prove to be superior to standard therapy for limited-disease patients awaits further study.
Recently, however, a randomized phase III study in extensive-disease patients compared cisplatin/irinotecan to the standard regimen of cisplatin/etoposide. The response rate was 89% with cisplatin/irinotecan; 67% with cisplatin/etoposide. Median survival and 1-year survival rate were 420 days and 60%, respectively, with cisplatin/irinotecan; 300 days and 40% with cisplatin/etoposide. The survival benefit was statistically significant (P = .0047; log-rank test).
Irinotecan has also been combined with etoposide in the salvage setting. Masuda et al treated 25 patients with relapsed or refractory small-cell carcinoma, all of whom had received prior platinum-based combination chemotherapy. Treatment was administered with G-CSF support. The major toxicities were myelosuppression and diarrhea. This highly active regimen achieved a 71% response rate and a median survival of 271 days.
Thus, the last several years have brought new chemotherapeutic agents with novel mechanisms of action into the therapy of small-cell carcinoma. Encouraging results obtained in both the salvage and extensive-disease settings have paved the way for using these new agents in the treatment of limited disease. The results of ongoing trials exploring various ways of combining or sequencing these agents with each other, with standard regimens, and with thoracic irradiation are anxiously awaited.