Abstract
Background/Aim: Neuroendocrine prostate cancer (NEPC) is rare and has a poor prognosis; its clinical course and treatment outcomes are also unclear. This study investigated the clinical characteristics, clinical course, and treatment outcomes of patients with NEPC. Patients and Methods: This retrospective study investigated 14 patients histologically diagnosed with NEPC at Kanazawa University Hospital between 2000 and 2019. Overall survival (OS) and progression-free survival (PFS) were retrospectively analyzed using the Kaplan–Meier method. Additionally, log-rank tests were used to compare survival distributions. Results: We included 14 patients histologically diagnosed with NEPC among 1,845 patients with prostate cancer. Four patients (0.22%) were diagnosed with de novo NEPC, and ten patients were diagnosed with NEPC during treatment. First-line platinum-based therapy’s objective response rate (ORR) was 66.7%, and disease control rate was 91.7%; median PFS was 7.5 months. The median OS from NEPC diagnosis was 20.3 months. The median OS of the liver metastasis (−) group was 31.6 months, and that of the (+) group was 9.4 months (p=0.03, hazard ratio=0.24). The median OS of the somatostatin receptor scintigraphy (SRS)-positive group was 31.6 months, and that of the SRS-negative group was 10.6 months (p=0.04, hazard ratio=0.14). Conclusion: Platinum-based chemotherapy is effective to some extent, but the duration of response is not sufficient; therefore, new treatment options are needed. This is the first study to show that SRS findings and the presence of liver metastases might be prognostic predictors of NEPC.
- Neuroendocrine prostate cancer
- small cell carcinoma
- somatostatin receptor scintigraphy
- liver metastasis
- prostate cancer
Prostate cancer (PC) is the most common cancer in men and the leading cause of cancer-related death in developed countries (1). Most PCs are adenocarcinomas, and neuroendocrine prostate cancer (NEPC) is rarely found (2). However, cases of castration-resistant PC (CRPC) progressing to NEPC, hereafter referred to as treatment-related NEPC (t-NEPC), have been reported recently (3, 4). NEPC is rarely detected as localized cancer, and more than 70%-75% of patients have distant metastases at the time of diagnosis, resulting in a poor prognosis (5–8). CRPC treatment includes alternative androgen deprivation therapy (ADT) (9), androgen receptor (AR) signaling-targeted agents (ARST) (10–13), and chemotherapeutic agents, such as docetaxel (DTX) (14, 15) and cabazitaxel (CBZ) (16, 17). Also, NEPC treatment is administered similarly to that for small cell lung cancer and is often treated with platinum-based chemotherapy (18–21). However, there are still few reports on chemotherapy efficacy for histologically diagnosed NEPC cases, and many aspects remain unclear. This study retrospectively reviewed the clinical characteristics, course, and treatment outcomes of 14 patients histologically diagnosed with NEPC.
Patients and Methods
This retrospective study included 14 patients histologically diagnosed with NEPC among 1,845 patients with PC treated at Kanazawa University Hospital from January 2000 to December 2019. Patients with NEPC included four diagnosed with NEPC at the time of prostate cancer diagnosis and ten treatment-related patients with NEPC (t-NEPC) diagnosed with NEPC during treatment. Patients with t-NEPC were diagnosed after undergoing rebiopsy or metastatic biopsy when NEPC was clinically suspected. For example, if there was exacerbation on imaging without prostate-specific antigen (PSA) elevation during treatment. Furthermore, we retrospectively reviewed the charts of all patients and analyzed relevant data. The collected medical data included age, serum PSA level, serum neuroendocrine tumor marker level, prostate biopsy pathology, clinical stage, and treatment progress. Clinical stage was determined based on the 2017 TNM Classification of Malignant Tumors, 8th Edition (22). Imaging studies, such as computed tomography (CT), magnetic resonance imaging, somatostatin receptor scintigraphy (SRS), and (18)F-fluorodeoxyglucose [(18)F-FDG] positron emission tomography/computed tomography (PET/CT) were performed during PC or NEPC diagnosis as well as at the time of disease progression. All decisions regarding intervals between subsequent imaging examinations and treatment strategies were left to the judgment of each attending physician. However, NEPC treatment was administered in accordance with the treatment for small cell lung cancer. The treatment regimens administered were EP (etoposide: 100 mg/m2 on days 1-3 and cisplatin: 80 mg/m2 on day 1 every 3 weeks), EC [etoposide: 100 mg/m2 on days 1-3 and carboplatin: area under the blood concentration time curve (AUC) 5 on day 1 every 3 weeks], irinotecan (100 mg/m2 on day 1, 8, and 15 every 5 weeks), IP (Irinotecan: 60 mg/m2 on day 1, 8, 15 and cisplatin: 60 mg/m2 on day 1 every 4 weeks), IC (Irinotecan: 60 mg/m2 on day 1, 8, and 15 and carboplatin: AUC 5 on day 1 every 4 weeks), amrubicin (40 mg/m2 on days 1-3 every 3 weeks), DTX (70 mg/m2 on day 1 every 3 weeks), everolimus (10 mg per day), and pembrolizumab (200 mg/m2 on day 1 every 3 weeks). Follow up was terminated on December 15, 2021. Survival was measured from NEPC diagnosis until death or the last follow up.
Overall survival (OS) and progression-free survival (PFS) were retrospectively analyzed using the Kaplan–Meier method. Progression was defined as the determination of PD according to Response Evaluation criteria in solid tumours (RECIST) 1.1. Log-rank tests were used for comparing the survival distributions, and statistical analyses were performed using Prism v.5 (GraphPad, San Diego, CA, USA). In all analyses, a p-value less than 0.05 indicated statistical significance. This study was approved by the Institutional Review Board of Kanazawa University Hospital (2016-328).
Results
Patient characteristics and treatment course of prostate adenocarcinoma (AC) are shown in Table I. Ten patients were diagnosed with pure AC at the initial prostate biopsy. The median age at AC diagnosis was 64.5 years (range=47-68 years). The median PSA level at the time of the first prostate biopsy was 43.7 ng/ml (range=3.1-2350 ng/ml), and all had Gleason score (GS) ≥8 AC, and 9 of 10 patients had metastatic disease. All patients underwent combined androgen blockade therapy as first-line treatment. After that, four patients received androgen receptor axis-targeting agents, and two received taxanes. The median response duration in the first-line treatment was 9.2 months.
Patient characteristics and treatment course of prostate adenocarcinoma.
Characteristics of patients with NEPC are shown in Table II. Of the 14 patients with NEPC, four (28.6%) were diagnosed with de novo NEPC, and ten patients with t-NEPC (71.4%) were diagnosed with NEPC during treatment. The median period from initial PC diagnosis to t-NEPC development was 11.1 months. Pathologically, six patients (42.9%) were diagnosed with pure small cell carcinoma (SmCC), and the remaining eight (57.1%) were diagnosed with mixed NEPC (SmCC or NEPC combined with AC). Ten patients (71.4%) had visceral metastasis at the time of NEPC diagnosis, eight (57.1%) had liver metastasis, and three (21.4%) had lung metastasis (including duplicated cases). Median PSA was 1.3 (range=0.04-176.8 ng/ml), median neuron-specific enolase (NSE) was 33.2 (range=10.4-400 ng/ml), and median pro-gastrin releasing peptide (Pro-GRP) was 57.1 (range=30.0-1,660 pg/ml). NSE was elevated in 11 patients (78.6%), and Pro-GRP was elevated in five patients (35.7%). SRS was performed in nine patients, and five (55.6%) were positive.
Characteristics of neuroendocrine prostate cancer patients.
Table III shows the results of immunohistological analysis of NEPC. The neuroendocrine markers, CgA, SNP, and CD56, were positive (including weak positive) in 11/14 (78.6%), 13/14 (92.9%), and 9/10 (90%) cases, respectively. AR and PSA were positive (including weakly positive) in only 1/5 (20.0%) and 1/6 (16.7%) cases, respectively (excluding mixed SmCC cases with positive or weakly positive AR or PSA).
Immunohistological analysis of neuroendocrine prostate cancer.
Clinical outcomes of neuroendocrine prostate carcinoma are shown in Table IV. As the first-line treatment for NEPC, EP therapy was performed in eight patients, EC therapy in three patients, IC therapy in one patient, and DTX in two patients. The first-line therapy’s objective response rate (ORR) was 57.1%, disease control rate (DCR) was 85.7%, and the median duration of response was 7.5 months. Patient No. 4, who received four EP courses followed by radiation therapy, achieved complete response (CR) and is still in CR. Second-line therapy included irinotecan in three patients, IC in two patients, EP in one patient, IP in one patient, everolimus in one patient, and CBZ in one patient. The ORR of the second-line therapy was 22.2%, DCR was 55.6%, and median duration of response was 3.9 months. Third-line therapy included amrubicin in two patients, EC rechallenge in one patient, and pembrolizumab in one patient. The median duration of response was only one month in all but one patient who received pembrolizumab. Patient No. 14, who had microsatellite instability-high (MSI-H), received pembrolizumab and is still in CR. Radiation therapy was administered in eight patients (57.1%).
Clinical outcomes of neuroendocrine prostate carcinoma.
The treatment effects of chemotherapy are shown in Figure 1. In Figure 1A, the waterfall plot shows the maximum change in the NE marker in first-line platinum-based chemotherapy. There were 11/12 (91.7%) patients with >50% NSE reduction with first-line platinum-based chemotherapy, and all patients had NSE reduction. More than 50% Pro-GRP reduction was observed in 6/12 patients (50%), and 9/12 patients (75%) had Pro-GRP reduction. PFS after first-line platinum-based chemotherapy is shown in Figure 1B. The ORR of first-line platinum-based therapy was 66.7%, DCR was 91.7%. The median PFS was 7.5 months. Figure 1C shows the waterfall plot of the maximum change in NE markers in second-line platinum-based chemotherapy. More than 50% NSE reduction was observed in 1/4 patients (25.0%) and 3/4 patients (75.0%) by second-line platinum-based chemotherapy. More than 50% Pro-GRP reduction was observed in 2/3 patients (66.7%), and 2/3 patients (66.7%) had Pro-GRP reduction.
Maximum change in NSE and Pro GRP due to chemotherapy and Kaplan–Meier curve for progression-free survival. (A) Maximum change in NSE and Pro GRP in 1st line platinum-based chemotherapy. (B) Kaplan–Meier curve for progression-free survival after 1st line platinum-based chemotherapy. (C) Maximum change in NSE and Pro GRP in 2nd line platinum-based chemotherapy. (D) Kaplan–Meier curve for progression-free survival after 2nd line platinum-based chemotherapy. (E) Maximum change in NSE and Pro GRP in 2nd line chemotherapy other than platinum-based. (F) Kaplan–Meier curve for progression-free survival after 2nd line chemotherapy other than platinum-based. NSE: Neuron specific enolase; Pro GRP: pro-gastrin-releasing peptide.
PFS after second-line platinum-based chemotherapy is shown in Figure 1D. The ORR of second-line platinum-based therapy was 50%, and DCR was 100%. The median PFS was 4.4 months. In Figure 1E, the waterfall plot shows the maximum change in NE marker in the second-line chemotherapy other than platinum-based. More than 50% NSE reduction was observed in 1/5 patients (20%) and 2/5 patients (40.0%) had NSE reduction by second-line chemotherapy other than platinum-based. Pro-GRP reduction was observed in one patient. PFS after second-line chemotherapy other than platinum-based is shown in Figure 1F. The median PFS was 1.5 months.
OS from NEPC diagnosis is shown in Figure 2. As shown in Figure 2A, the median OS was 20.3 months. OS for de novo NEPC and t-NEPC is shown in Figure 2B. The median OS of the de novo NEPC group was 26.7 months, and that of the t-NEPC group was 11.7 months. OS according to liver metastasis is shown in Figure 2C.
Overall survival from NEPC diagnosis. (A) Overall survival from NEPC diagnosis. (B) Overall survival for de novo NEPC and t-NEPC. (C) Overall survival according to liver metastasis. (D) Overall survival according to SRS findings. NEPC: Neuroendocrine prostate cancer; SRS: somatostatin receptor.
The median OS of the liver metastasis (–) group was 31.6 months, and that of the liver metastasis (+) group was 9.4 months. OS was significantly shorter in the liver metastasis (+) group (p=0.03, hazard ratio=0.24). OS according to SRS findings is shown in Figure 2D. The median OS of the SRS-positive group was 31.6 months, and that of the SRS-negative group was 10.6 months. The median OS was significantly longer in the SRS-positive group (p=0.04, hazard ratio=0.14).
Discussion
De novo NEPC is very rare, and its incidence is reported to be less than 1%-2% (23, 24). In a retrospective study from the Surveillance, Epidemiology, and End Results database of the National Cancer Institute, 510/560,124 patients with PC had de novo NEPC with an incidence of 0.09% (25). In this study, the incidence of de novo NEPC was 4/1845 (0.22%), which was similar to previous reports. The incidence of t-NEPC has been reported to be increasing in recent years (4). In a prospective study of mCRPC patients, t-NEPC was found in 27/160 patients (16.9%) (26). In this study, the median time from PC diagnosis to the onset of t-NEPC was 13.8 months, similar to 20 months in a review of 123 patients with t-NEPC (6). NEPC is AR-independent, and NSE and Pro-GRP are considered helpful as serum NE markers (27–29). In this study, most patients had high NSE and Pro-GRP levels. However, 3/14 patients (21.4%) were histologically diagnosed as NEPC but had normal levels of both NSE and Pro-GRP. In a systematic review, 22.8% of t-NEPC cases were diagnosed based on morphological criteria only, suggesting that it is crucial to diagnose histologically even when NE markers are negative.
The median OS of NEPC (including de novo NEPC and t-NEPC) was reported to be 7-16.8 months, compared to 20.3 months in this study (6, 20, 30–33). In previous reports, the ORR for platinum-based first-line chemotherapy was 34%-77.8%, with a PFS of 3.8-10.1 months (20, 30–34). The ORR of platinum-based first-line chemotherapy was relatively good at 66.7%, but the median PFS was short at 7.5 months in this study. Although platinum-based chemotherapy is effective to some extent, the duration of response is not sufficient, and new treatment options are needed. Recently, a phase II trial with the Aurora kinase A (AURKA) inhibitor, alisertib, was presented with a median PFS of 8.7 weeks. Although it failed to meet its primary endpoint with a 6-month PFS of only 12.6%, 3/59 patients showed exceptional remissions or disease stabilizations (35). In a preliminary retrospective analysis of seven patients treated with the mammalian target of rapamycin (mTOR) inhibitor, everolimus, a decrease in tumor markers was noted in five patients (36). Recently, immune checkpoint inhibitors (ICIs) received approval as a treatment option for patients with advanced small cell lung cancer (SCLC). Although ICIs did not dramatically improve SCLC prognosis, it is possible that increasing the therapeutic options may result in improvements in t-NEPC prognosis if ICI therapy becomes an approved treatment strategy for t-NEPC in the future. A phase III trial was conducted to add programmed death-ligand 1 (PD-L1) inhibitor to EP or EC for advanced small cell lung cancer, which significantly increased OS in the additional PD-L1 inhibitor group (37, 38). Several clinical trials of PD-L1 inhibitors for NEPC are underway, and their results are awaited (NCT04848337, NCT03179410, NCT04709276, NCT03582475). Pembrolizumab, a PD-L1 inhibitor, has received accelerated approval from the US Food and Drug Administration for treating solid tumors with high microsatellite instability (MSI-H) or mismatch repair deficiency (dMMR), regardless of primary tumor site (39). MSI-H or dMMR is found in 2%-12% of metastatic PC cases and should be considered in NEPC cases (40). In this study, MSI-H was observed in patients with progressive disease (PD) after three EC therapy cycles and two irinotecan cycles. The patient was treated with pembrolizumab, which resulted in CR, and is currently on cycle 7 and maintaining CR.
A PARP inhibitor, olaparib, is effective in patients with metastatic CRPC with breast cancer susceptibility (BRCA) mutations (41). It has been reported that about 12% of patients with metastatic CRPC have BRCA mutations, making it an option for NEPC (42).
SCLC is sensitive to radiation therapy (RT), and concurrent chemoradiation is the standard of care for limited-stage SCLC (43). Radiotherapy is effective in cases of NEPC (6, 34). In this study, we found a case of CR after four cycles of EP followed by radiation therapy in a patient with NEPC without distant metastasis. RT may be an effective treatment option for NEPC without distant metastasis.
The prognostic factors for NEPC are unknown. Somatostatin receptors are more abundantly expressed in neuroendocrine tumors (NETs) than normal tissues (44). SRS has been performed as a standard test for pancreatic and gastrointestinal NET worldwide because of its high detection sensitivity (45, 46). Also, SRS has been reported to be useful in NEPC (47, 48). Also, its accumulation was the highest in NET G1 in the revised World Health Organization classification of endocrine organs, 2019, with weaker SRS accumulation in NET G3 (49). Furthermore, a prospective study of 38 patients with NETs showed that SRS-positive patients had a significantly better prognosis than SRS-negative patients (50). In a retrospective study of 79 patients with pancreatic NET, the prognosis was significantly better with a stronger expression of somatostatin receptor-2a by immunostaining (51). In this study, the median OS of SRS-positive patients was 31.6 months, and that of SRS-negative patients was 10.6 months, indicating a significantly longer OS in SRS-positive patients. To the best of our knowledge, this is the first study in the world to show that SRS findings may predict prognosis in NEPC. In addition, OS was significantly worse in the liver metastasis (+) group, as shown in this study. Furthermore, SRS and liver metastases may be prognostic factors in NEPC.
There are several limitations to this study. First, this study was a retrospective study of a small number of patients. Furthermore, the treatment intervals, imaging evaluation, and the need for rebiopsy and metastatic site biopsy were left to the attending physician’s judgment.
Conclusion
We reviewed the clinical features and treatment outcomes of histologically confirmed NEPC. Platinum-based chemotherapy is effective to some extent, but the duration of response is not sufficient, and new treatment options are needed. This study showed for the first time that SRS findings and the presence of liver metastases might be prognostic predictors of NEPC. Further prospective and large-scale studies are needed to confirm the results of this study.
Footnotes
Authors’ Contributions
H.I. designed the experiments. H.I., R.N., T.M., T.K., S.K., H.Y., S.K., K.I, and Y.K. collected clinical data. H.I., R.N., T.M., S.K., K.I. and A.M. analyzed the data. H.I., K.I., and A.M. drafted and revised the manuscript. All Authors read and approved the final version of the manuscript.
Conflicts of Interest
All Authors declare that there are no potential conflicts of interest relevant to this article.
- Received February 12, 2022.
- Revision received February 25, 2022.
- Accepted February 28, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.