Small cell neuroendocrine carcinoma of the cervix: From molecular basis to therapeutic advances

Small cell neuroendocrine carcinoma of the cervix (SCNECC) is an uncommon but aggressive uterine malignancy, the cause of which is generally associated with human papillomavirus (HPV) infection. A lack of clinical trials and evidence-based treatment guidelines poses therapeutic challenges to this rare tumor. At present, published data remain limited to case series and case reports. While clinical management has traditionally followed those of small cell neuroendocrine (SCNE) lung cancer relying on surgery, chemoradiation, and systemic chemotherapy, the prognosis remains dismal. Immune checkpoint inhibitors (ICIs), such as monoclonal antibodies that target programmed death-1 (PD-1) or programmed death-ligand 1 (PD-L1), atezolizumab and durvalumab have proven effective in extensive-stage SCNE lung cancer. Moreover, pembrolizumab has also proven beneficial effects when added onto chemotherapy in metastatic and recurrent HPV-associated non-SCNE cervical cancer. It holds promise to use ICIs in combination with chemoradiation to improve the clinical outcomes of patients with SCNECC. Future advances in our understanding of SCNECC biology – associated with the study of its genomic and molecular aberrations as well as knowledge from SCNE of lung and other extrapulmonary sites– would be helpful in discovering new molecular targets for drug development. Collaborative efforts and establishment of a SCNECC-specific biobank will be essential to achieve this goal.


Introduction
High-grade neuroendocrine tumors (NETs) represent a highly heterogeneous group of neoplasms derived from ubiquitous neuroendocrine cells, most commonly arising from the lungs, the gastrointestinal tract, and the pancreas [1].NETs may occasionally develop outside of their most common primary tumor sites, including the female genital tract.Small cell neuroendocrine carcinoma of the cervix (SCNECC) is an aggressive form of NET that affects the uterine cervix [2e4].It is rare and accounts for less than 2% of all invasive cervical malignancies.Unfortunately, the prognosis of SCNECC remains poor because of its frequent recurrence and the high metastatic burden at diagnosis [5].Histopathologically, the tumor is composed of round to spindle-shaped cells with an infiltrative growth pattern [6].Hyperchromatic nuclei, lymphovascular space invasion (LVSI), and necrotic areas are common findings.Immunohistochemical markers of neuroendocrine differentiation e including chromogranin, CD56, and synaptophysin e are usually, but not invariably, positive [7,8].In addition to the unfavorable outcomes [9], the rarity of the disease poses major therapeutic challenges due to a shortage of clinical trials and lack of evidence-based treatment guidelines.The clinical experience described in case series and case reports still represents the only available source of information of this disease.
In this narrative review, we present an overview of the pathogenesis of SCNECC and elaborate the clinical management of patients with this rare cervical malignancy.Our discussion further summarizes the present knowledgement and advancement in this field.We further delve into the updates on the tumorgenesis of high-grade NETs, especially focusing on pulmonary and extrapulmonary small cell neuroendocrine (SCNE) neoplasms.The objective is to derive strategies that can be potentially applicable to SCNECC.The following sections aim to highlight the aspects that are most relevant to improving our understanding of SCNECC tumorigenesis, thereby developing innovative and effective treatment.

Role of human papillomavirus infection
The majority of SCNECC can be linked to oncogenic human papillomavirus (HPV) infections, with HPV18 being the predominant genotype [6,10e12].A comprehensive systematic review and meta-analysis of 403 cases revealed that 85% of SCNECC samples were tested positive for HPV [11].Specifically, following the integration of the viral genome into infected host cells, the main viral oncoproteins, E6 and E7, sequestered the cellular ubiquitin-proteasome system to promote rapid degradation of the tumor suppressor proteins TP53 (tumor protein p53) and Rb1 (RB transcriptional corepressor 1), respectively [13].Furthermore, the E6 protein may drive tumorigenesis through its ability to activate the expression of the catalytic subunit of telomerase e an enzyme activity which is crucial for immortalization [13].In addition to binding to Rb, the E7 protein can promote malignant transformation by driving cell cycle progression and disrupting the epigenetic control of normal cellular lifespan through its ability to bind histone deacetylases [13].The Cancer Genome Atlas Research Network demonstrated that HPV-associated cervical carcinomas exhibit several significantly mutated genes (SMGs) such as SHKBP1 (SH3KBP1 binding protein 1), ERRB3, CASP8 (caspase 8), HLA-A (major histocompatibility complex, class I, A), and TGFR2 (transforming growth factor beta receptor 2).These SMGs also include newly identified ones like the mutations driven by the apolipoprotein B mRNA editing catalytic polypeptide-like driven mutations, however, no SCNECC was included [14].

Surgery or chemoradiation?
Advanced stage, lymph node metastasis, deep stromal invasion, not receiving chemotherapy, and large tumor size are related to prognosis [27e33].However, clinical outcomes for SCNECC patients tend to vary based on the treatment approach.The challenge of treating patients with SCNECC is featured by the lack of information from randomized clinical trials to guide the selection of optimal therapies [27].According to the National Comprehensive Cancer Network guidelines, patients with stage I ( 4 cm) SCNECC can be treated with radical hysterectomy and chemotherapy or chemoradiation [28].However, treatment guidelines for advanced and/or recurrent disease are lacking, and data from a small number of retrospective studies remain the only source to inform patient management [29e33].At present, the mainstay of treatment includes surgery, radiotherapy, and systemic chemotherapy (Table 2).A study conducted by the Taiwanese Gynecologic Oncology Group (TGOG) [29] enrolled 179 patients with SCNECC.Of the 144 with early-stage disease, 110 underwent primary surgery and 34 received primary radiotherapy.Patients in the latter group (6%) showed a significant reduced rate (p ¼ 0.009) of loco-regional failure compared with the former (27%), suggesting that radiotherapy with chemotherapy may yield superior outcomes in terms of loco-regional control.Moreover, in a specific subset of 13 patients with a tumor size <2 cm and no LVSI underwent radical surgery, their 5-year overall survival (OS) was as high as 89% (2 of the 4 without adjuvant chemotherapy recurred and died of disease) [29].Excluding the 13 patients with excellent prognosis, the 5year OS rate was significantly better in the 14 patients underwent at least 5 cycles of platinum-based chemothrapy and primary radiotherapy.The OS rate for this group was significantly higher (78%) compared to the 46% in the 97patients (p ¼ 0.046) [29].For advanced disease, the TGOG study analysed patients with stages IIBÀIVB.The results showed that chemoradiation with platinum and etoposide (PE), if completed at least five cycles significantly improved both the 5-year failurefree survival (62.5% versus 13.1%, respectively; p ¼ 0.025) and cancer-specific survival (CSS) (75.0%versus 16.9%, respectively; p ¼ 0.016) [31].Unfortunately, the 5-year CSS of 0% was observed for patients with stage IVB [31].
On analyzing a pooled analysis of 188 patients with stages IÀIVB SCNECC, Cohen et al. [30] conducted a pool analysis of 188 patients with stage I-IVB SCNECC.They found that the 5year OS of patients who received radical hysterectomy (38.2%) was higher than those of women who did not receive the operation (23.8%) (p < 0.001).This reflects that pateints treated Table 1 Comparison of genetic features of high-grade NECC, more common cervical carcinomas, and SCNE lung carcinomas.

Genetic alterations
High-grade NECC a (n ¼ 97) [23] SCC, AD, ASC [14] SCNE lung carcinoma (n ¼ 1800) [  with surgery were at a lower disease or had smaller tumors.In addition, Cohen et al. [30] reported the 5-year CSS rates of 9.8% and 0% in stage IIBÀIVA and IVB SCNECC, respectively.Importantly, patients with stage IIBÀIVA disease who received chemotherapy (as primary adjuvant or concurrent with radiotherapy) showed a better 3-year OS rate of 17.8% as opposed to the 12% in those who did not receive chemotherapy (p ¼ 0.04).
A Thai study highlighted that early-stage SCNECC patients who underwent primary surgery and adjuvant chemotherapy had better survival than those who did not receive adjuvant therapy or adjuvant radiotherapy/chemoradiation [32].Furthermore, the authors reported that older age (>60 year old) and deep stromal invasion were significant poor prognostic factors [32].A Korean multi-center study (n ¼ 102) found that parametrial involvement and LVSI were associated with unfavorable prognosis [33].A study from China found that stage and tumor size were significant independent prognostic factors for early stage SCNECC [34].Additional study substantiates the finding that postoperative chemotherapy can improve the OS of patients diagnosed with early-stage SCNECC, compared to those who undergo primary surgery alone [35].Postsurgical chemotherapy regimens have been compared, and the combinations of PE, as well as vincristine, doxorubicin, and cyclophosphamide (VAC), outperforms other treatment regimens [36,37].Collectively, these observations suggest that tumor size and deep stromal invasion may improve the prognostic stratification of SCNECC in early-stage disease.Moreover, PE chemotherapy has demonstrated an improvement in OS for patients undergoing primary surgery or radiotherapy for stage I-IV disease.

Role of neoadjuvant chemotherapy
Data on the potential utility of neoadjuvant chemotherapy (NACT) in SCNECC remain inconclusive [38,39].A study by Dongol et al. [38] described four women with stage IB1ÀIIB SCNECC who received NACT followed by radical hysterectomy.Unfortunately, only one patient was alive when the study was published.In a Japanese multi-center study of 126 SCNECC patients, it was found that fewer than 4 chemotherapy cycles, paraaortic lymph node metastasis, locally advanced disease and distant metastasis were all associated with a worse outcome [40].Furthermore, the response rate to PE therapy was found to be superior compared to the taxaneplatinum regimen (43.8% vs 12.9%) [40].Conversely, Lee et al. [39] indicated a worse prognosis for women with stage IB-IIA SCNECC who underwent NACT followed by radical surgery.This study also suggested that the burden of residual diseases following NACT may have a prognostic significance.Accordingly, Bermudez et al. [41] showed that the 5-year OS of women with residual tumors smaller than 2 cm was significantly higher than for those with tumors larger than 2 cm (58% versus 21%, respectively; p < 0.05).Wang et al. noted a trend towards worse failure-free survival (41.2% vs 60.5%, p ¼ 0.086) in patients who received NACT before surgery as compared to those received primary surgery with or without adjuvant therapy [31].In summary, a wide range of therapeutic regimens have been attempted in the treatment of SCNECC, resulting in a multimodality approach [28].Generally, patients with earlystage (stage IB1) tumors are recommended to undergo radical hysterectomy followed by adjuvant chemotherapy or chemoradiation.However, primary radical hysterectomy can be reserved for patients whose tumors are less than 2 cm without deep stromal invasion/LVSI, as they can achieve excellent outcome with just postoperative adjuvant chemotherapy, thereby avoiding the need for adjuvant chemoradiation.For patients with more advanced disease, definitive chemoradiation could be an option.As far as chemotherapy is concerned, PE should be currently considered as the most effective regimen.Nonetheless, the role of NACT in SCNECC remains unclear.

Potential utility of immunotherapy
Recently, ICIs have demonstrated significant improvements in progression-free survival (PFS) in various solid tumors [42e44].Programmed death-1 (PD-1), a negative co-receptor found in antigen-stimulated B and T cells, has a crucial role in mediating cytotoxic effects on cancer cells [45].Pembrolizumab (Keytruda®) e a potent humanized immunoglobulin G4 monoclonal antibody e has high binding affinity to the PD-1 receptor and inhibits its interaction with both PD-L1 and programmed cell death ligand 2 (PD-L2).Pembrolizumab has an acceptable preclinical safety profile and is currently as an intravenous immunotherapy for advanced malignancies.
Research has found that HPV-associated cervical cancer expressed increased levels of PD-L1 [46], specifically, HPV16 E7 has the ability to directly induce PD-L1 expression [47].This has led to the testing of pembrolizumab as a potential treatment for patients with cervical cancer [48].The phase Ib KEYNOTE-028 trial demonstrated a 17% overall response rate for PD-L1-positive cervical cancer patients [49].Furthermore, the KEYNOTE-158 trial indicated that pembrolizumab monotherapy produced a durable antitumor activity with manageable safety profile in patients with advanced cervical cancer [49].The comprehensive evaluation from the KEYNOTE-826 trial, which involve 617 patients with recurrent, persistent, or metastatic cervical cancer.The trial reported that pembrolizumab significantly outperformed placebo plus platinum-based chemotherapy in terms of PFS, having 10.4 months versus 8.2 months, respectively (hazard ratio [HR] ¼ 0.65; 95% confidence interval ¼ 0.53À0.79;p < 0.001) [42] (Table 3).While the three trials mentioned did not encompass patients with SCNECC, the role of HPV infection in the pathogenesis of this rare malignancy has prompted additional clinical testing of pembrolizumab.The potential of incorporating ICIs into chemoradiotherapy as a front-line treatment for locally advanced cervical cancer is currently being explored [50,51].Regrettably, a phase II trial on the use of pembrolizumab monotherapy in recurrent SCNECC was discontinued due to an unsatisfactory response [52,53].Nonetheless, a patient with stage IVB SCNECC (with bone metastasis and liver metastasis) achieved long-term survival benefits from combined pembrolizumab and b i o m e d i c a l j o u r n a l 4 6 ( 2 0 2 3 ) 1 0 0 6 3 3 chemoradiotherapy [54].Another PD-1 inhibitor, nivolumab, demonstrated therapeutic effect in a patient with large-cell neuroendocrine cervical cancer when coupled with radiotherapy [55].A patient with metastatic SCNECC who showed a complete response following treatment with nivolumab and radiotherapy [56].However, despite these promising outcomes, we believe that targeting PD-1/PD-L1 e particularly utilizing ICIs monotherapy e may not suffice to ensure an adequate control of SCNECC after failure of first-line treatment.
Notwithstanding, the potential utility of pembrolizumab as an add-on to curative chemoradiation warrants further exploration.For example, data from the KEYNOTE-604 study highlighted that the addition of pembrolizumab to EP as firstline therapy improved both PFS (HR ¼ 0.75, p ¼ 0.0023) and OS in patients with extensive SCLC (HR ¼ 0.8, p ¼ 0.0164) [57].Anti-PD-L1, atezolizumab and durvalumab, have also proven efficacy in extensive-stage SCLC as front-line therapy (atezolizumab: HR of OS 0.70, p ¼ 0.007 and durvalumab HR of OS 0.73, p ¼ 0.0047) [58,59].These findings suggest that the addition of ICIs to current chemotherapy regimens could enhance therapeutic effects against SCNECC.More studies specifically focusing on SCNECC are necessary to conclusively determine the potential benefits of ICIs.

Molecular subtypes and targeted therapies in SCNE carcinoma of lung and other extrapulmonary sites
Recent studies have discovered molecular subtypes of SCNE lung cancer defined by ASCL1 (achaete-scute family bHLH transcription factor 1), NEUROD1 (neurogenic differentiation 1), POU2F3 (POU class 2 homeobox 3), and YAP1 (Yes1 associated transcriptional regulator) transcription factors [60,61].Targeted agents are actively tested in SCLC and a range of solid tumors (Table 4) [60,61].Lurbinectedin, an inhibitor of DNA active transcription of protein-coding genes showed high overall response rate (35.2%) in SCNE lung cancer [62], is currently undergoing a randomized trial [63].Promoters of ASCL1-and NEUROD1-dependent genes were found as specific targets of lurbinectedin in SCNE lung cancer cells [64].
Agents targeting a homologous recombination deficiency, poly (ADP-ribose) polymerase (PARP) inhibitors such as olaparib and rucaparib, as well as ATR kinase inhibitor and Wee1 inhibitors are currently under investigation for their potential in treating SCNE lung cancer (Table 4).Combination of topoisomerase 1 inhibitor, VX-970 (M6620), depleted ataxia-telangiectasiaemutated and rad3-related protein kinase, may stop  cancer cells from preventing the repair of DNA damaged by chemotherapy [65].Among 21 patients with metastatic or unresectable malignancies (comprising five cases of SCNE lung cancer and one case of SCNECC), who were treated with an ATR inhibitor (M6620), three out of the five patients with platinum-refractory SCNE lung cancer had a partial response or prolonged stable disease, lasting 10, S 6, and 7 months, respectively [65].However, the phase III trial which was investigating the use of VX-970 was unfortunately suspended (NCT03896503).Meanwhile, the results from an umbrella trial for recurrent SCNE lung cancer were reported [66] Interestingly, Schlafen 11(SLFN11), a protein that acts parallel to homologous recombination pathway, is a biomarker to guide therapy selection in PARP inhibitors [67].Patients who test SLFN11 positive may be eligible for atezolizumab with or without talazoparib (NCT04334941).Furthermore, EZH2 depletion has been shown to potentiate MYC degradation inhibiting tumor growth in xenograft of small cell carcinoma [68].EZH2 inhibitor (DS-3201 b), an epigenetic modifier to reverse SLFN11 promoter methylation, is being investigated for recurrent SCNE lung cancer (NCT03879798).

Conclusions and future directions
The high metastastic potential of SCNECC still poses formidable challenges.Previous therapeutic approaches to this malignancy showed suboptimal efficacy.Our present knowledge on the mechanisms of SCNECC tumorigenesis remains limited and the low quality of evidence has hampered the development of recommendations for clinical management.For now, incorporation of PE chemotherapy in the management of SCNECC of all stages is recommended.Furthermore, ICIs such as anti-PD1/PD-L1, in combination with PE chemotherapy, have proven efficacy in extensive-stage SCNE lung cancer as front-line therapy and recurrent, persistent, or metastatic cervical cancer.Thus, SCNECC patients not in the best prognostic group (<2 cm, no LVSI/deep stromal invasion) may be offered add-on ICIs out of clinical trials.Several shortcomings of published studies in the field should be considered, including 1) small sample sizes due to the rarity of SCNECC, 2) lack of detailed information on its molecular underpinnings, and 3) the inherent biases of non-randomized study designs.In the future, two main research directions await investigation.First, new therapeutic approaches to SCNECC may be informed by positive results obtained in other NETs, and may thus receive specific scrutiny in SCNECC.Second, a comprehensive dissection of the underlying molecular aberrations, including genetic and epigenetic drivers, will be a key prerequisite to enhance therapeutic outcomes.When coupled with collaborative biobanking, which is paramount to achieve sufficiently large samples in the field of rare malignancies, it is anticipated that high-throughput, integrative "omics" approaches supported by advanced bioinformatics tools will permit to uncover exhaustive biological information on SCNECC that will offer substantial support for future tailored treatments.
b i o m e d i c a l j o u r n a l 4 6 ( 2 0 2 3 ) 1 0 0 6 3 3

Table 2
Summary of published studies (sample size: >50 cases) on treatment approaches and clinical outcomes of patients with small cell neuroendocrine carcinoma of Abbreviations: CCRT-EP5þ: concurrent radiotherapy and at least 5 cycles of etoposide and platinum chemotherapy; CSS: cancer-specific survival; FFS: failure-free survival; CT: chemotherapy; EP5þ: at least five cycles of etoposide and platinum; NACT: neoadjuvant chemotherapy; OS: overall survival; Pt: patient; RT: radiotherapy; S: surgery; vs: versus; TTP: time to progression.

Table 3
Overview of clinical trials using immune checkpoint inhibitors in recurrent/advanced cervical cancer.

Table 4
Overview of clinical trials on targeted therapies in small cell neuroendocrine carcinoma of the lung or other anatomical sites.