The specific aims of our retrospective study were to elucidate and evaluate the efficacy and tolerability of monotherapy with TMZ as second and third line treatment after platin-based chemotherapy in patients with NECs (WHO G3). Usually these patients have a very poor prognosis with a short-term survival. In our material, none of the 28 patients had objective response (CR + PR) to treatment, but a disease control rate of 38%. The OS and PFS after start of TMZ treatment were 3.5 months and 2.4 months, respectively. The first report on second and third line TMZ-based chemotherapy in 25 patients with NECs after progression on first line chemotherapy (cisplatin/etoposide or docetaxel and doxorubicin) [5
] showed overall response rate of 33% and disease stabilization in 38% of the patients. The median PFS was 6 months and the median OS was 22 months. The patients were treated with TMZ alone (n
= 5) or in combination with capecitabine (n
= 19) of which a subgroup also had bevacizumab (n
= 7). Adding capecitabine and bevacizumab to TMZ did not seem to have any additional effect. However, the number of patients in each group was small [5
], which may hide a beneficial effect of combined therapy. These data are indicative of a promising effect of TMZ in NECs; however, data may not be comparable to our study due to differences in the selection of patients. We found a median Ki-67 proliferation index of 50% in our patients, whereas half of the cohort in the other study [5
] had a Ki-67 at 20–30%, which may account for the difference in the present results. Although not obvious in the recent study [5
], we had a high inclusion (21%) of patients with PS 2, which may affect the results.
TMZ and capecitabine-based regimes are associated with relatively high tumor response rates in patients with well or moderately differentiated pancreatic NENs (G1 and G2) [4
]. We showed a trend towards longer median overall survival for patients with primary pancreatic NECs (7.0 months) versus patients with nonpancreatic NECs (2.9 months) after receiving TMZ as monotherapy. The PFS was 3.3 months versus 1.9 months, respectively. Although different NEN populations, the results are supported by a phase 2 study of 29 patients (28 well-differentiated neuroendocrine tumors and one poorly differentiated neuroendocrine carcinoma) treated with TMZ and thalidomide [8
]. The report demonstrated radiologic response rate of 25%—therapy appeared to be most active among metastatic pancreatic NENs with a response rate of 45% [8
The use of TMZ with capecitabine as first line therapy for pancreatic NENs G1 + G2 has been reported with a response rate of 70% [6
]. If these data are consistent and confirmed, TMZ-based chemotherapy regimens with relatively mild toxicity may develop into a future second line or even first line option in these patients.
The role of the nuclear antigen Ki-67 as a prognostic indicator and a surrogate marker for a therapeutic response is still a matter of debate. A review from 2008 [9
] endorses Ki-67 immunostaining, particularly in pancreatic NENs with reference to the WHO classification that supports Ki-67 immunostaining as a routine in the immunohistochemical examination of NENs. Our subgroup analysis showed a significantly shorter median survival in patients with a Ki-67 index ≥50% (). The study of TMZ-based chemotherapy of NECs [5
] also found more responders in patients with Ki-67 <60% than in those with higher Ki-67. This may reflect the difference in tumor biology between tumors with high and low Ki-67, respectively, and thus call for an extended division into subgroups than the current WHO 2010 classification. In contrast to the previous study [5
], positive or negative CgA immunohistochemistry or positive or negative SRS were not predictive factors for survival, which may be due to the small sample size.
-methylguanine DNA methyltransferase (MGMT) is a DNA repair enzyme that is believed to induce cancer cell resistance to O6
-alkylating agents, for example, TMZ [10
]. This is, however, not a consistent finding, as other studies found no significant difference in the response rate after TMZ-based regimens when correlated with the MGMT expression [5
]. The role of MGMT promoter methylation status as a predictive factor when it comes to NEC patients treated with TMZ is not consistent and was not a part of our retrospective analysis. To look further into the hypothesis of MGMT expression and TMZ response, a study of a greater cohort than ours is needed.
In conclusion, TMZ is an option for palliative treatment in patients with NECs as the toxicity is acceptable, and treatment may be received on an outpatient basis. Overall TMZ had limited effect in the second and third line treatment of patients with NECs in the present study. However, our data indicate that a subset of patients with pancreatic NECs may benefit from the treatment. Further and larger studies of TMZ as a second line treatment of NEC are warranted, particularly in pancreatic NECs and probably in the form of combination chemotherapy. In order to optimize the result of the investigation, it is presupposed that the patients are very carefully selected having high performance status and low Ki-67 indices.