Abstract
Background: Extrapulmonary small cell carcinoma (EPSCC) is a rare disease of highly proliferating neuroendocrine tumor cells. Therapy and outcome of EPSCC patients treated at our institution were retrospectively analyzed. Patients and Methods: The records of all patients diagnosed with EPSCCs between 1998 and 2007 at the Charité Berlin were retrieved. Stage of disease, therapy, treatment response, time to progression and overall survival were assessed. Results: Twenty-eight patients with EPSCC were identified. Primary tumor sites were genitourinary tract, head and neck, gastrointestinal tract, breast, and unknown primary. Fifteen patients had locoregional (LD) and 13 had distant metastatic disease (ED). Median survival was 43 and 12 months in LD and ED patients. Four out of seven LD patients receiving chemotherapy plus local treatment remained free of disease. Conclusion: Prolonged survival can be achieved in patients with LD by multimodal therapy. Overall survival in patients with ED is comparable to patients with ED small cell lung cancer.
Extrapulmonary small cell carcinomas (EPSCC) can arise in multiple organ sites and form a rare clinicopathological entity of highly proliferative epithelial neuroendocrine tumors with aggressive biological behaviour (1). The incidence is estimated to be 0.1% to 0.4% (2), and is expected to be underestimated since some cases might be misdiagnosed as small cell carcinoma of the lung (SCLC), which shares many clinicopathological features with EPSCC (3). Since its first description in 1930 (4), a variety of extrapulmonary primary sites have been described including head and neck, esophagus, stomach, pancreas, gall bladder, colon, kidney, urinary bladder, prostate, and uterine cervix (5-10).
The limited data and the variety of primary sites make choice of treatment of these tumors difficult. Due to the biological relatedness of EPSCCs to high-grade neuroendocrine tumors of the lung, concepts for staging and therapy of the latter were often applied (8, 9, 11).
The aim of our study was to review all cases at our institution during the past ten years. The goal was to investigate efficacy of chemotherapy in advanced stages and to define whether combined modality treatment including chemoradiotherapy derived from limited disease SCLC management was associated with favourable outcome in locoregional disease.
Patients and Methods
Selection of patients. We retrieved all clinical records from patients with EPSCC treated at Charité, Berlin between 1998 and 2007. At the time of diagnosis, patients had to have no evidence of pulmonary disease on computed tomography (CT) scan of the lungs and/or bronchoscopy. Histopathological criteria for diagnosis was in correspondence with the WHO classification of lung neoplasias (12). Mixed histologies were accepted, but a growth fraction of at least 50% in the neuroendocrine differentiated cell compartments was required for inclusion. Patients with large cell carcinomas and Merkel cell tumors were excluded.
Data collection and evaluation. Demographics, primary site, past medical history, ECOG performance status, stage, treatment modalities and clinical response, time to progression (TTP), and overall survival (OS) were assessed retrospectively. Metastatic disease without identifiable primary site was classified as ‘carcinoma of unknown primary’ (CUP). According to the Veterans' Administration Lung Study Group (VALSG) system for SCLC, EPSCC were classified as ‘limited disease’ (LD), if the tumor was localized in a single organ with or without locoregional lymph node involvement or ‘extensive disease’ (ED), in case of metastatic disease outside locoregional lymph nodes. Clinical response (complete response (CR), partial response (PR), stable disease (SD), progressive disease (PD)) was classified according to RECIST criteria (13). Cumulative time to progression (TTP) and overall survival (OS) were estimated with the Kaplan-Meier method, subgroups were compared using the log-rank test. Because of multiple testing, all p-values provided have to be regarded as exploratory.
Results
Patient characteristics. A total of 28 patients with EPSCC with follow-up information were identified. Sixteen patients were male, 12 were female, their age ranged from 23 to 78 years, with a median of 62. Fifteen patients were diagnosed with LD, 13 had ED. Median follow-up was 42 month. Thirteen patients had a primary in the genitourinary tract, of which uterine cervix, bladder, and prostate were the primary site in four patients, respectively, and one patient had a primary of the vagina. One patient had an EPSCC of the breast. Other primary sites were paranasal sinus in four cases, larynx in two cases, gastrointestinal tract in four cases, and in four cases the site of the primary was unknown (Table I).
Of four patients with EPSCC of the prostate, two did not have evidence of adenocarcinoma of the prostate gland, one patient showed a mixture of focal EPSCC and glandular adenocarcinoma, and one had a neuroendocrine small cell carcinoma with glandular growth pattern coexpressing prostate-specific antigen (PSA) and chromogranin A. One of the patients with SCC of the prostate had an adenocarcinoma treated with hormone deprivation before, which was not detectable in biopsies at time of diagnosis of the SCC. All three patients with SCC of unknown primary and liver metastases had an active adenocarcinoma of the prostate without histological evidence for neuroendocrine differentiation in the prostate.
Treatment and outcome of patients with limited disease (LD). Stage, primary, treatment modalities, and outcome are summarized in Table II. Fifteen patients had locoregional disease at the time of first diagnosis. Eleven underwent primary resection, two received primary chemoradiotherapy, one had chemotherapy only and one had androgen deprivation therapy plus radiotherapy for the primary in the prostate gland.
Seven patients received two or three treatment modalities including chemotherapy. Four of these patients remained relapse free after 21, 24, 64, and 68 months, whereas the three other patients relapsed 7, 11, and 21 months after diagnosis.
Seven patients received locoregional therapy only. Four patients were treated with surgery plus radiotherapy; one remained free of disease after 26 month, the others relapsed after 5, 12, and 95 months. All 3 patients treated with surgery or radiotherapy alone relapsed and died of disease (Table II).
The median TTP and OS of patients with LD were 21 (95% confidence interval (95% CI) 0-46) and 43 months (95% CI 9-77), respectively. Two-year OS was 64% (Figure 1). OS of LD patients treated with multimodal therapy was superior to patients treated with local or systemic therapy only (two-year survival rates 83% and 42%, five-year survival rates 63% and 21%, respectively). The median OS of the group receiving multimodality treatment was not reached at the time of analysis, whereas the median OS of patients treated locally or systemically was 22 months (95% CI 0-49, Figure 2).
Treatment and outcome of patients with extensive disease (ED). Twelve out of the thirteen patients with ED received systemic chemotherapy resulting in a response rate of 58%. Eleven patients relapsed and received second line chemotherapy with only one single objective response. Median TTP was 5 months (95 % CI 1-8) and median OS was 12 months (95% CI 6-15). Compared to LD patients, TTP and OS was significantly shorter in ED patients (Figure 1).
Discussion
This study retrospectively assessed the treatment and outcome of 15 patients with LD and 13 patients with ED extrapulmonary neuroendocrine small cell carcinoma.
The histogenesis of EPSCC is far better understood today than it was a few years ago. Initially it was believed that tumorgenesis of EPSCC was based on the malignant transformation of ubiquitous cells of the ‘amine-precursor uptake and decarboxylase’ (APUD) system. This hypothesis offered no explanation for the often described co-existence of a small cell neuroendocrine carcinoma with histological areas of an adenocarcinoma or squamous cell carcinoma of the primary organ site within a single malignant lesion (5, 14-16). There is increasing evidence that EPSCC can arise from pluripotent basilar cells capable of divergent differentiation with the consequence that the neuroendocrine phenotype arises as a transdifferentiation phenomenon in progression of an organ-specific carcinoma (15, 17, 18). In our study, the mixed histologies of adenocarcinoma and neuroendocrine carcinoma and especially a case with coexpression of PSA and chromogranin A within the same cell population supports the notion of a common background of both histotypes.
Because EPSCC is a rare entity, only retrospective series have been reported. A single center analysis from Korea in 24 patients with various primary sites of disease concluded that EPSCC is usually a fatal disease with a discouraging outcome for various treatment modalities, except for a subgroup of patients with LD and a primary of the uterine cervix (7). In contrast, our analysis demonstrated a favorable outcome in patients with locoregional disease regardless of the organ site of the primary. Long-term survival was achieved by multimodal treatment including surgery plus chemo- or chemoradiotherapy with the exception of a single patient remaining disease free for 26 months after surgery and adjuvant radiotherapy. Therefore, our analysis supports the concept of multimodal therapy, as has become standard for limited disease SCLC (19). Concordant to results in ED in SCLC, response rates to chemotherapy and survival of ED patients in our study were poor (20, 21).
- Received March 25, 2009.
- Revision received May 27, 2009.
- Accepted June 9, 2009.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved