Due to the rarity of NECCs, primarily retrospective studies have reported on outcomes. Unlike squamous cell carcinoma, which is often highly curable with loco regional therapy, NECC has poor outcomes, even when diagnosed at an early stage, suggesting the need for improvement in both primary and salvage therapies in NECC [5,6,7]. Expression of somatostatins in neuroendocrine tumours of other subsites is well established and there have been previous studies, both for diagnosis and for treatment with peptide receptor radionuclide therapy (PRRT). The sensitivity and specificity of Ga-DOTATATE PET in the detection of NETs are reported as 96% and 100%, respectively [11, 12]. In our study, no patients had undergone Ga-PET CT-based imaging.

For NECC diagnosed in early stage (FIGO I–IIA), the Society of Gynecologic Oncology and Gynecologic Cancer Intergroup recommends that patients should undergo radical hysterectomy with lymphadenectomy along with systemic chemotherapy in either neo adjuvant or adjuvant setting [13, 14]. Published literature reports that in patients with early-stage disease, the 5-year overall survival is 71% [15]. In our study, the 5-year overall survival for early-stage was 68.4%. The role of post-surgical adjuvant treatment in patients who have already received systemic chemotherapy is not fully described in the literature. Recently published meta-analysis on 13 studies that used adjuvant radiation in patients who underwent radical hysterectomy for NECC cervix has shown a benefit in locoregional control, with pelvic recurrence reported in 24.3% of patients not receiving adjuvant radiation vs. 12.5% in those receiving adjuvant radiation (p 0.09) [16]. There was, however, no improvement in overall survival. Small studies have also advocated the use of systemic chemotherapy before definitive local therapy because of high distant failure rates. Neoadjuvant chemotherapy may help in sterilizing systemic micrometastasis, but it may lead to prolongation of overall treatment time and accelerated tumour repopulation compromising local control. Chang et al., in a phase II study, evaluated neoadjuvant chemotherapy in early-stage NECC. The study investigators alternated 2–3 cycles of vincristine, adriamycin and cyclophosphamide with cisplatin and etoposide (VAC/PE) before hysterectomy and reported good response with neoadjuvant chemotherapy, but on hysterectomy microscopic residual tumour was present in all, emphasizing the need for local treatment [17].

For locally advanced NECC, there is no uniform chemotherapy regimen that is recommended or uniformly utilized in clinical practice. However, cisplatin/carboplatin with etoposide is the most commonly used treatment followed by cisplatin/carboplatin and paclitaxel [6]. For locally advanced NECC, the Gynecologic Cancer InterGroup (GCIG) Consensus Review recommends chemoradiation or systemic chemotherapy with cisplatin and etoposide in combination with local treatment [13]. The National Cancer Database for patients with NECC demonstrated that patients who received brachytherapy along with radiotherapy in locally advanced NECC had significantly improved survival than those patients who received only external radiation or no radiation at all [18]. In our group of patients, 49/56 patients (87.5%) received external radiation either in a radical or adjuvant setting, and all of these patients also received brachytherapy.

The most common pattern of relapse in our study was distant metastasis, occurring in 73% of patients. In our study, we specifically investigated the patterns of relapse to determine the need if any for prophylactic cranial irradiation in NECC. Unlike small cell lung cancer, the role of prophylactic cranial irradiation (PCI) is not proved or tested in NECC of cervix cancer. Based on patterns of relapse in this cohort, we noted that in NECC of the cervix, brain metastasis was uncommon, while the most common first site of metastatic presentation was liver, lung, and bone. Viswanathan et al. studied 21 patients of NECC and no patient had brain metastases as the sole site of first recurrence, though two patients developed brain metastases subsequently with lung metastases [19].

The survival outcomes have been dismal in neuroendocrine tumours with the majority of patients presenting with visceral metastasis. Peptide receptor therapy has been used with Lutetium-177 (177Lu)-DOTATE for the treatment of neuroendocrine tumours primarily for non-cervical neuroendocrine tumours. It is considered highly effective in advanced or metastatic neuroendocrine tumours. These therapeutic agents that are utilized for lung and gastro-pancreatic neuroendocrine tumours need further investigation in neuroendocrine cervical cancer as no prospective studies integrating DOTATOC/DOTATATE are available in gynaecological tumours [20, 21].

Novel therapeutics such as immune checkpoint inhibitors and targeted therapies could possibly be beneficial for these tumours. Immune checkpoint inhibitors are a promising group of drugs in neuroendocrine neoplasms. They are currently being tested in a phase II study of NEC of thoracic, pancreatic and gastrointestinal origin who have progressed on prior treatment [22, 23]. There are no prospective trials recruiting patients with NECC. Three case reports have been reported in the literature on the use of immune checkpoint inhibitors in NECC. In two of these, nivolumab (programmed death-1 inhibitor, PD-1 checkpoint inhibitor) has been used at recurrence, with only partial response [24,25,26]. The authors demonstrated a near complete resolution of tumour for > 10 months. In another case report, a patient with recurrent NECC received MEK inhibitor, trametinib outside the clinical trial and the patient was reported to be disease-free at 8 months [26].

In a recent study, 40 pathological samples from patients with NECC underwent IHC testing for the following markers: mismatch repair proteins (MMR, MLH1, MSH2, MSH6, PMS2 on 28 samples), PDL-1 (31 samples), somatostatin (19 samples) and poly ADP-ribose polymerase (PARP, 11 samples) on formalin-fixed paraffin tissue. All 28 (100%) samples showed intact impressions for mismatch repair proteins. In the 31 samples tested for PD-L1 expression, 26 (84%) tested negative for the receptor, for somatostatin expression, 18/19 (95%) were negative, while for PARP, 10/11 (91%) samples showed positive PARP expression, potentially suggesting that further trials may be conducted with PARP inhibitors [27].

The finding of low PD-L1 expression was also corroborated by a phase II study conducted on 7 patients with recurrent lower genital tract neuroendocrine cancer, of which 6 were NECC and one patient had a vulvar neuroendocrine tumour. All patients received pembrolizumab at a dose of 200 mg 3 weekly. Median progression-free survival was only 2.1 months. The authors demonstrated that pembrolizumab alone shows only minimal activity in NECC [28].

In our study, we conclude that patients with locally advanced NECC who did not receive systemic chemotherapy in addition to chemoradiation did worse than those who also received systemic chemotherapy. Our analysis has shown that outcomes for NECC have remained dismal with multimodality approach. Limitations of our study include the retrospective nature of study, and since we report outcomes of patients with NECC spanning over more than a decade, there has been heterogeneity in practice for both early- and locally advanced-stage NECC. Also, the mode of radiotherapy treatment and dose fractionation have changed over time. Thus, it is not plausible to make definite statements based on impact of radiation and brachytherapy dose. We also could not capture the response to radiation at the time of brachytherapy.

In conclusion, NECC is a tumour with a poor prognosis and innovative treatment options are recommended with a combination of surgery, radical radiation with neoadjuvant or adjuvant chemotherapy with cisplatin and etoposide. Though multiple treatment options have been explored, such as peptide-based receptor therapy, addition of immunotherapy and MEK inhibitors, the outcomes for NECC have remained dismal with pooled 5-year overall survival of 34% [6]. Due to increased expression of PARP in tumour cells, further trials with addition of PARP inhibitors may be conducted to ascertain any improvement in outcomes. Given that NECCs are rare tumours, prospective data pooling across various institutions may be needed to understand the impact of systemic vs. intensified concurrent chemotherapy. Furthermore, prospective multi-centric rare tumour registries will be beneficial in assessing the long-term outcomes of patients with this rare histology, especially with differential IHC expression in tumours.