Objective: Combination chemotherapy with the two agents streptozotocin (SZT), which is a nitrosurea, and 5-fluorouracil (5-FU), an alkylating agent, has a long-established role in the treatment of neuroendocrine tumours; however, it is often accompanied by considerable toxicity, and it has not been assessed in a comparative manner with other current chemotherapy regimens. In order to assess the therapeutic response and adverse effects using an alternative nitrosurea, lomustine (CCNU), which has a different side-effect profile, in combination with 5-FU, we have reviewed all patients with neuroendocrine tumours who received this form of treatment in our department.
Design: Retrospective analysis of the case notes of patients with metastatic neuroendocrine tumours who received treatment with the combination of CCNU and 5-FU, and who were followed up according to a defined protocol in a given time frame.
Patients: Thirty-one patients with metastatic neuroendocrine tumours (18 with carcinoid tumours, five islet-cell tumours, five chromaffin-cell tumours and three medullary carcinoma of the thyroid) treated with the combination of CCNU and 5-FU, and when necessary additional therapy, over a 22-year period, were included in this analysis.
Measurements: The symptomatic, hormonal and tumoural responses before and after chemotherapy with the combination of CCNU and 5-FU over a median follow-up duration of 25 months (range 9-348 months) were recorded. Of the 31 patients (16 males; median age 52 years, range 20-86 years), eight (four males; median age 61 years, range 30-74 years) were treated with the combination of CCNU and 5-FU alone (Group 1), whereas the other 23 patients (12 males; median age 47 years, range 20-86 years) received additional therapy with other chemotherapeutic regimens, somatostatin analogues, alpha-interferon or radiolabelled meta-iodobenzylguanidine (131I-MIBG) therapy (Group 2).
Results: A total of 121 therapeutic cycles was administered (mean 3.9, range 1-14 cycles). None of the patients obtained a complete tumour response. A partial tumour response (not a complete but a 50% or greater reduction of all measurable tumour) was seen in six out of the 29 patients (21%) (four out of eight in Group 1 and two out of 21 in Group 2, respectively). There was no tumour progression in eight out of the 29 patients (27.5%) (one out of eight in Group 1 and seven out of 21 in Group 2, respectively). The median survival over the period of the study was 48 months (95% confidence interval, CI, 22-74 months). The overall 5-year survival rate was 42% (95% CI, 17-67%) for all patients and 50% (95% CI, 18-83%) for the carcinoid group alone, according to Kaplan-Meier analysis. A complete or partial symptomatic response was obtained in 12 out of 27 (44%) patients who presented with symptoms (four out of eight in Group 1 and eight out 19 in Group 2, respectively) and a complete or partial hormonal response in eight out of 19 patients (42.1%) who presented with hormonally active disease (two out of four in Group 1 and six out of 15 in Group 2, respectively). Nine out of the 15 (60%) patients with carcinoid tumours who presented with symptoms obtained a symptomatic response, five out of 10 patients (50%) a hormonal response, and four out of 16 (25%) patients a partial tumoural response, respectively. The combination of CCNU and 5-FU was safe and well tolerated. Serious side-effects necessitating the termination of CCNU and 5-FU were seen only in two patients, and mainly consisted of reversible bone marrow suppression. No chemotherapy-related death was recorded.
Conclusions: Chemotherapy with CCNU and 5-FU, either alone or in combination with other therapeutic modalities, produces considerable symptomatic and hormonal improvement and moderate tumour regression/stabilization according to currently accepted WHO criteria, particularly in patients with metastatic gastroenteropancreatic neuroendocrine tumours with minimal adverse effects. However, long-term survival was still relatively poor. It may therefore be a valuable additional therapl was still relatively poor. It may therefore be a valuable additional therapeutic option, particularly for well-differentiated carcinoid and islet-cell tumours, but mainly reserved for when there is no response or progression of the disease after currently available first-line treatment with somatostatin analogues or radiopharmaceuticals.