Clinical Investigation

Clear cell carcinoma of the uterine cervix; an unusual HPV-independent tumor: Clinicopathological features, PD-L1 expression, and mismatch repair protein deficiency status of 16 cases


  • Pınar Bulutay
  • Özgür Can Eren
  • Özlem Özen
  • Asuman Nihan Haberal
  • Nilgün Kapucuoğlu

Received Date: 22.04.2023 Accepted Date: 01.06.2023 Turk J Obstet Gynecol 2023;20(3):164-173 PMID: 37667475


Endocervical clear cell carcinoma (c-CCC) is a rare and HPV-independent adenocarcinoma type of cervix. Being usually resistant to conventional chemotherapy. Immunotherapy has recently been added as a preferred regimen as a second-line treatment option for programed cell death-ligand 1 (PD-L1)-positive or mismatch repair (MMR) deficient cervical carcinomas. In this study, clinicopathological features, PD-L1 expression, and MMR deficiency status of c-CCCs were investigated.

Materials and Methods:

Sixteen c-CCC diagnosed cases were included in this study. PD-L1 expression was evaluated using two different PD-L1 clones (22C3 and SP263). MMR deficiency status of the cases was evaluated using four MMR proteins (MLH1, PMS2, MSH2, and MSH6).


Most of the c-CCC cases were presented as FIGO Stage I (68.75%). PD-L1 expression in either tumoral or tumor-infiltrating immune cells (TILs) was present in 62.5% (10/16) and 69% (11/16) of the 22C3 and SP263 clones, respectively. Most of the cases with high TIL density were also positive for PD-L1. The PD-L1 expression rate was less than 50% in most of the cases and 12.5% of the cases shared extensive PD-L1 staining. Overall, MMR deficiency was observed in 31.25% of the cases. Most of the MMR-deficient cases (80%) were PD-L1 positive.


Although our study cohort is limited, we have shown that PD-L1 expression and MMR deficiency can be found in c-CCCs in variable degrees. These findings suggest that accompanying TIL density and MMR deficiency could be used as candidates for predicting PD-L1 positivity for c-CCCs. However, to indicate the clinical importance of these findings, objective treatment outcomes of cases treated with immunotherapy should be seen.

Keywords: Endocervical clear cell carcinoma, PD-L1 22C3, mismatch repair deficiency

PRECIS: The presence of PD-L1 expression and Microsatellite Instability in endocervical clear cell carcinomas predicts the immunotherapy may yield promising results in the treatment of endocervical clear cell carcinomas as well.


Clear cell carcinoma (CCC) of Müllerian origin is a rare tumor with distinct histology that may occur in the ovary, vagina, uterus and cervix. Among carcinomas of the cervix, the most frequent histologic type is squamous cell carcinoma (SCC), which represents 75% of all cases. This is followed by endocervical adenocarcinoma (ECA), which accounts for 20-25% of all cases. ECA in fact represents a heterogeneous group of tumors with various etiologies, molecular drivers, morphologies, responses to treatment, and prognoses. ECA classification has recently been reorganized by the International Endocervical Adenocarcinoma Criteria and Classification and updated by the 2020 World Health Organization with separation into human papillomavirus (HPV)-associated (HPVA) and HPV-independent (HPVI) categories(1). Cervical clear cell carcinoma (c-CCC) is one of the HPVI ECA types and accounts for only 3.3% of all ECAs(2). Although the clinicopathological features of other Müllerian system-derived CCCs have been broadly studied, little is known about the clinicopathological features and optimal treatment strategies of c-CCCs due to their rarity.

C-CCCs is morphologically identical to their endometrial and ovarian counterparts, with solid, tubulocystic, and papillary architectures. Tumor cells are typically characterized by clear cytoplasm and hobnail nuclei, prominent cell membranes, hyperchromatic nuclei, and low mitotic rate. Oxyphilic, flat, and signet ring cells can be seen albeit rarely. Clear cells are round or polyhedral and contain abundant glycogen and occasionally hyaline globules(3,4).

The prognosis of c-CCC varies with stage, but the actual risk associated with this histology is unknown. Although most studies on ECA survival rates have not evaluated c-CCC separately, they have shown that HPVI ECAs have a worse prognosis than HPVAs(5).

The treatment approach c-CCC is consistent with the other cervical cancer types. Radical hysterectomy or trachelectomy, pelvic lymphadenectomy, +/- external beam radiotherapy (EBRT), and +/- brachytherapy constitute a standard treatment approach for early-stage cervical carcinoma (FIGO stages IA and IB1), while EBRT and systemic chemotherapy are the standard surgical treatment regimen for further stages (FIGO stages IB2, II, and III). According to the NCNN (National Comprehensive Cancer Network) version 1.2023 cervical cancer guidelines, pembrolizumab has been added as a preferred regimen as a second-line option for treating programed cell death-ligand 1 (PD-L1) positive or microsatellite instability-high (MSI-H)/mismatch-repair deficient (dMMR) tumors(6,7,8). However, PD-L1 expression status in c-CCCs has been demonstrated in only a few studies in the English literature, and a successful immunotherapy response has recently been published as a case report(9,10,11).

As is known, four commercial PD-L1 expression assays linked to different PD-1/PD-L1 checkpoint inhibitors are currently available for the treatment of several cancer types  22C3, 28-8, SP142, and SP263. Among these, the 22C3 assay has received FDA approval as a “companion diagnostics” for the treatment of pembrolizumab in many cancer types, including cervical cancer(12).

Furthermore, MSI-H/dMMR is identified as a biomarker for immunotherapy efficacy and pointed to the potential use of immune checkpoint inhibitors in several cancer types including cervical cancer by the Keynote-158 trial(13). PD-L1 expression, MSI status, and their correlation with clinicalopathologic features have already been investigated in other HPVA and HPVI-type cervical carcinomas in larger series(14). However, the data regarding the situation for these biomarkers in c-CCCs are limited. In this study, we investigated the prevalence of PD-L1 expression (by using PD-L1 22C3 and SP263 assays) and MMR (by using MLH1, MSH2, MSH6, and PMS2 proteins) deficiency status and their relationship with the clinicalopathological features in our c-CCC series.

Materials and Methods

Case Selection and Clinicopathological Evaluation

A total of 16 primary c-CCC cases were included in this study. Among them, 14 cases were selected from the 105 ECA cases that were gathered previously for the design of the reproducibility of the new ECA classification study by our team(5). After adding 2 new cases to expand the cohort, all available hematoxylin and eosin-stained (H&E) slides were reviewed by two pathologists to confirm the diagnosis and determine the optimal tumor-containing tissue block.

Diagnostic confirmation was made by identification of the classic morphologic features of c-CCC, including high-grade tumor cells with hobnail nuclei and prominent nucleoli in solid, papillary, and/or tubulocystic architectures. Napsin-A, p53, ER, PR immunohistochemical stains, and HPV-DNA in situ hybridization (ISH) techniques were also applied for the confirmation of the morphologic diagnosis The presence of stromal tumor-infiltrating lymphocytes (TILs) was assessed independently on each slide. Stromal TILs are reported as a percentage of tumor stroma occupied by lymphocytes. TILs are classified into three groups using the following cut-off values: 10% (mild), 10-40% (moderate), and >40% (high).

Clinicopathological parameters, including age at diagnosis, presence of lymphovascular space invasion, nodal status, and International Federation of Gynecology and Obstetrics (FIGO) stage, were recorded from electronic medical records. It was also confirmed that the tumors originated from the endocervix by confirming that there was no tumor infiltration either in the ovary or in the endometrium.

Immunohistochemical Assessment

The 4-um-thick whole tissue sections were taken from formalin-fixed paraffin-embedded blocks. PD-L1 immunohistochemistry was conducted with two PD-L1 antibody clones on two different staining platforms. The SP263 antibody clone (Roche Diagnostics, Mannheim, Germany) was used on the Ventana Benchmark Ultra platform, and the 22C3 antibody clone (Agilent Technologies, Waldbronn, Germany) was run on a DAKO Autostainer Link 48 at the Koç University Hospital, Pathology Department (Istanbul, Turkey). All assays are referred to hereafter by the antibody clone used.

The combined positive score (CPS) and tumor proportion score (TPS) were used for evaluating PD-L1 positivity. CPS was calculated by dividing the total number of viable tumor cells by the number of cells stained with PD-L1 (including tumor cells, lymphocytes, and macrophages) and multiplying by 100. Only intra- and peritumoral immune cells were counted for scoring immune cells in the CPS system. Stromal immune cells from outside the tumor were not included. The percentage of viable tumor cells with partial or complete membrane staining at any intensity was used to calculate TPS. Cut-off score 1 was considered positive for CPS, and over 50 was considered as extensive staining(9,15). For each stain, including PD-L1, appropriate positive and negative controls were included.

MSI status was evaluated by immunohistochemistry using DNA mismatch repair (MMR) proteins; MLH1 (1:200, Abcam, Cambridge, England), MSH2 (1:50, Roche, Mannheim, Germany), MSH6 (1:100, Roche, Mannheim, Germany), and PMS2 (1:100, Abcam, Cambridge, England). Simultaneous expression of four MMR proteins (MLH1, MSH2, MSH6, and PMS2) was considered “proficient DNA mismatch repair (pMMR)”. Otherwise, “deficient DNA mismatch repair (dMMR)” is defined as the absence of at least one of the four indexes stated above. The normal expression was defined as nuclear staining within tumor cells, with expression in tumor-infiltrating lymphocytes as the positive internal control. The absence of nuclear staining within tumor cells despite concurrent positive labeling in internal nonneoplastic tissues was described as a loss of expression.

Napsin-A (mouse monoclonal, MSVA-112), estrogen receptor (ER, 1:50, monoclonal rabbit ab, clone SP1), progesterone receptor (PR, 1:50, monoclonal rabbit ab, clone 1E2), p53 (1:200, monoclonal mouse ab, clone DO7), and p16 (prediluted, monoclonal mouse ab, clone E6H4) were stained on an automatic immunostainer [Ventana Benchmark XT (Ventana Medical Systems, Tucson, AZ, USA)]. ER and PR stains were scored on a continuous quantitative scale based on the percentage of nuclear staining in tumor cells (0-100%). Focal or diffuse granular cytoplasmic staining was recorded as positive for Napsin-A. Diffuse, block-like staining of moderate or strong intensity was accepted as positive for p16, while patchy or no staining was interpreted as negative. p53 was noted as mutated either in the complete absence (null pattern) of staining or in the strong staining of >75% of tumor cell nuclei. The other positive staining rates were accepted as wild-type for p53(16). The non-HPV associated status of the cases was already shown in a previous study using the HPV-DNA ISH technique (Detailed knowledge about the HPV-DNA ISH technique can be found in the referenced study)(5).

Immunohistochemical stains were initially independently assessed by two pathologists (P.B., O.C.E.). In the case of ambiguity, a consensus diagnosis was reached with a gynecopathologist with 20 years of experience in multi-head microscope (N.K.).

Statistical Analysis

To correlate PD-L1 expression with MMR status, clinicopathological features, and different TIL group, evaluations were made via 2-tailed χ2 tests. For every analysis, statistical significance was set with p-value <0.05. IBM SPSS Statistics for Windows version 28.0 (IBM Corp., Armonk, NY, USA) was used for the analysis.


Clinicopathological characteristics

The clinical characteristics of the cases are summarized in Table 1. The median age at diagnosis of 16 cases was 54 years (range 31-79 years), and 18% of the cases were under 40. There was no history of in utero diethylstilbestrol (DES) exposure in any of the patients. Most cases presented with vaginal bleeding and/or watery discharge. All patients underwent radical hysterectomy with bilateral salpingo-oophorectomy and pelvic-para-aortic lymph node dissection. The mean tumor size was 3.8 cm (1.0-8.0 cm). Three of the tumors invaded the upper two-thirds of the vagina (18.75%, 3/16) and one invaded the left parametrium (6.25%, 1/16). Two penetrates the vaginal surgical margin, one extended to the posterior cervical margin, and one concurrently had a tumor implant in the serosa of the sigmoid colon. The remaining tumors are confined to the cervix with a negative surgical margin (75%, 12/16). The initial FIGO stages for most patients were stages I and III (87.5%), except for one patient with stage IIA1 and one with stage IVA. The initial FIGO stages were distributed as follows: stage IA1: 12.5% (2/16), IB1: 6.25% (1/16), IB2: 18.75% (3/16), IB3: 31.25% (5/16), IIA1: 6.25% (1/16), IIIC1: 6.25% (1/16), IIIC2: 12.5% (2/16), and IVA: 6.25% (1/16). Lymphovascular invasion (LVI) was  in eight (50%) cases and absent in eight (50%) cases. LVI was not seen in the early-stage tumors (FIGO Stages IA and IB1). Lymph node metastasis (LNM) at the time of diagnosis was  in 4 (25%) and absent in 12 (75%). Two of these were pelvics (50%) and two were paraaortic (50%) lymph nodes. Survival data of 14 cases were available. The mean follow-up time of the cases was 36 months. Two of the cases died on the 13th and 18th months after surgery (Cases are numbered 1 and 10, respectively, according to their position in Table 2). Other cases were still alive with no recurrence or metastasis in the following period.

Histologically, 6 (37.5%) tumors had an exophytic polypoid appearance with mostly superficial infiltration of the endocervical mucosa. One case was restricted to the endocervical epithelium, with a polypoid appearance (5 cm in size) and no evident cervical stormal invasion. Fifteen cases showed Silva pattern-C infiltration (93.75%), and one case had a Silva pattern-B infiltration as there was no significant cervical stormal invasion (6.25%)(17).

The histological appearance of the cases was similar to other gynecological system origins, with an ordinary CCC appearance (Figure 1A). The tumor cells were arranged in a tubullocystic, papillary, or solid architecture (often a mix to varying degrees) and were surrounded by cells with clear (intracytoplasmic glycogen), eosinophilic, granular, and sometimes hobnailed cytoplasm with minimal stratification. In some cases, pseudonuclear inclusions and hyaline globules were seen. One case consisted of well-differentiated cysts lined by a single flattened epithelial layer on a hyalinized stroma that could easily be mistaken for benign entities as defined. TILs were seen in all cases at different rates (Figure 1B-D). However, stromal high TIL was observed in 7 (43.75%) cases. Moderate and mild TIL was observed in 18.75% and 37.5% of cases, respectively. Carcinoma in situ foci defined for CCCs in the literature were not observed in any of the cases(18). Immunohistochemically, all tumors were Napsin-A positive and wild-type with p53. Conveniently, ER, PR, or P16 expression was not seen in any case.

Mismatch Repair Protein Deficiency

Thirty-one percent (5/16) of the c-CCC cases demonstrated MMR deficiency (Table 2). Two of them showed a dual loss of MSH2 and MSH6 (Figure 2), 1 showed a dual loss of MLH1 and PMS2, 1 showed a triple loss of MLH1, PMS2, and MSH6, and 1 showed a single loss of MSH6 proteins. Forty percent (2/5) of dMMR c-CCC cases showed extensive PD-L1 positivity in both TPS and CPS scores. PD-L1 positivity was seen in 80% of dMMR cases. Eighty percent (4/5) of dMMR cases were accompanied by a high rate of TILs. The remaining dMMR case exhibited no PD-L1 expression in tumor cells or TILs. The relationship between PD-L1 expression and MMR status was not statistically significant for either tumor (p=0.3) or combined tumor and inflammatory cells (p=1) (Table 3).

Tumoral and Peritumoral Immune PD-L1 Expression

PD-L1 expression status and their relationship with the MMR results for each case are shown in Table 2. In 69% (11/16) of cases with SP263 clones and 62.5% (10/16) of cases with 22C3 clones, either the tumor (TPS) or immune cells like lymphocytes and macrophages that were infiltrating the tumor (CPS) were stained with PD-L1. PD-L1 expression was seen in 56.25% (9/16) of the cases based on TPS in both clones. Both PD-L1 SP263 and 22C3 clones showed  perfect accordance with TPS. For CPS also, concordance was excellent, except for one case that was considered negative for the SP263 clone (which was 5% positive for the 22C3 clone) (Table 2). Two of the cases showed extensive PD-L1 staining in both clones (18%; 2/11) (Figure 3). Most of the remaining PD-L1-positive cases showed tumoral staining in less than 50% of cells (82%; 9/11). The mean PD-L1 expressing rates for CPS were 16% for SP263 and 15.4% for 22C3. PD-L1 positivity was seen only in TILs in 2 cases (18%; 2/11), with one where positivity was seen only on the SP263 clone (5% PD-L1 staining rate).

Relationship Between PD-L1 Expression and Clinicopathological Features

The relationship between PD-L1 expression and clinicopathological parameters of the patients is shown in Table 3.

There was no correlation between CPS/TPS with age (p=0.61/p=0.11), tumor size (p=0.13/p=0.59), lymphovascular invasion (p=0.60/p=0.61) or lymph node metastasis (p=0.60/p=0.77) based on PD-L1 expression. There was no correlation between PD-L1 expression and FIGO stage as well (p=0.60/p=0.83). PD-L1 expression was higher in both TPS and CPS with high amounts of TIL. However, statistical significance was not found (CPS, p=0.061, TPS, p=0.35).


Immune checkpoints such as programmed cell death 1 (PD-1) and its ligand (PD-L1) are critical in antitumor immunity, and blocking them has been demonstrated to enhance outcomes in patients with numerous types of malignancies(19). Pembrolizumab, a PD-1 inhibitor, has been approved for the treatment of patients with PD-L1-positive cervical cancer that is locally progressed, recurrent, or metastatic(20). In this study, a objective was to evaluate PD-L1 expression in c-CCC.

The CCC of the cervix accounts for approximately 3.3% of ECAs, which is further reduced if SCCs are also included in the cohort. Therefore, there is a paucity of data in the literature regarding the prognostic indicators, clinical outcomes, and treatment strategies of c-CCC. According to recent studies, the prognosis of early-stage c-CCC is similar to that of other types of cervical cancer. Advanced-stage and lymphatic involvement are associated with worse survival for both progression-free survival (PFS) and overall survival (OS) for c-CCC(21,22,23,24,25). According to Liu et al.(24), the 5-year OS for FIGO stages IB to IIA and stage IIB to IIIC was 95.7% and 46.2%, respectively. Thomaset al.(25) have similarly found that the presence of positive lymph nodes has a negative impact on 5-year PFS (31% vs. 92%, p<0.001) and 5-year OS (80% vs. 100%, p=0.02) in stage I and IIA c-CCC patients. Stolnicu et al.(26) have recently compared the survival outcome between c-CCC and ECA and found a significant difference in 5- and 10-year OS between c-CCC and HPVA ECA, whereas no significant difference was shown between c-CCCs and gastric-type ECAs. Moreover, they emphasize the importance of the stage in OS with their results; They stated that OS in stage I CCC was 85.3% at both 5 and 10 years, while it was 39.7% at 5 years and 0% at 10 years in stages II to IV (p<0.001)(26).

Despite variations in survival and treatment responses between cervical HPVA and HPVI carcinomas, there is currently no therapeutic difference between SCCs and adenocarcinomas, including c-CCCs(27,28). Radical surgery combined with targeted adjuvant therapy may cure early-stage disease. However, more advanced diseases are treated with EBRT and systemic chemotherapy(6). According to the NCNN Version 1.2023 guidelines for cervical cancer treatment, concurrent chemoradiation is generally the primary treatment choice for stages of IB3 to IVA disease. In 2020, the FDA approved the addition of the PD-1 inhibitor pembrolizumab to the treatment of PD-L1-positive (CPS1)- or dMMR patients whose disease progresses after chemotherap(29,30,31). Keynote-158 (pembrolizumab)(13), Empower-Cervical-1 (cemiplimab)(32), and Keynote-826 (pembrolizumab)(7) were pivotal studies that indicated immunotherapy enhanced overall survival in both post-platinum failure and frontline persistent, recurrent, or metastatic PD-L1 positive cervical cancer patients.

According to the Keynote-158 trial, the overall response rate was 14.4% in PD-L1-positive patients. Median PFS and OS for PD-L1-positive patients were 2.1 and 11 months, respectively. However, no responses were observed in patients with PD-L1-negative tumors. Regarding safety, 4.1% of patients stopped treatment because of treatment-related adverse events (including hepatitis, severe skin reactions, and adrenal insufficiency)(13). The Empower-Cervical-1 study investigated the therapeutic efficacy of cemiplimab (PD-1 inhibitor) in 608 patients (304 of the patients randomly received cemiplimab and 304 received chemotherapy) with recurrent or metastatic cervical cancer. Although only a small portion of their patients could have their PD-L1 expression assessed, cemiplimab had a longer median overall survival than the chemotherapy group. (13.9 vs. 9.3 months) among the PD-L1 positive patients (≥1%). The median overall survival rates were 7.7 and 6.7 months with cemiplimab and chemotherapy in PD-L1-negative patients, respectively. According to the results, objective responses to cemiplimab were observed in 18% (≥1%) of PD-L1 positive patients and 11% (<1%) of PD-L1 negative patients. In the overall population, an objective response was obtained in 16.4% of patients in the complimab group compared with 6.3% in the chemotherapy group. This means that PD-L1-positive patients generally have an increased overall survival benefit. However, it can be concluded that PD-L1-negative patients also have an overall survival benefit with cemiplimab as or slightly better than patients receiving chemotherapy. Of their complimab and chemotherapy-received cohorts, 45% and 53.4% had grade 3 or higher adverse events, and 15.7% and 0.7% had immune-related adverse events, respectively(32).

The relative benefit of adding pembrolizumab (PD-1 inhibitor) to platinum-based chemotherapy with or without bevacizumab in PD-L1-positive metastatic or unresectable cervical cancer patients was investigated in 548 patients by Keynote-826 (pembrolizumab) study(7). According to their results, progression-free (10.4 vs. 8.2 months) and overall survival (24-month estimate of patients alive, 53.0% vs. 41.7%) were significantly longer with pembrolizumab than with the placebo group. According to their study’s adverse event data, 42.4% of patients in the placebo group and 49.8% of patients who received pembrolizumab experienced major adverse events. Only hypothyroidism (18.2% vs. 9.1%) and a lower white blood cell count (12.1% vs. 7.1%) posed a greater risk in the pembrolizumab group.

PD-L1 protein expression is currently used as a predictive biomarker for checkpoint therapy in cervical cancer. However, the heterogeneous expression tendency of the PD-L1 protein makes this method suboptimal. Therefore, the PD-L1 protein expression rate may not be directly associated with prognostic significance and treatment response. The patient selection according to PD-L1 protein expression excludes potential patients for whom checkpoint therapy could be effective. Rotman et al.(33) investigated the tumoral PD-L1 expression heterogeneity for cervical cancer. According to their results, 27% of cases had heterogeneity between different tumor cores based on the percentage of positive tumor cells. Additionally, for comparison, they also applied the RNAish technique and observed heterogeneity in 11% of the cases. Their results showed that core biopsies can consequently lead to false negative results and the RNAish technique could serve as a better biomarker than IHC detection.

In this study, the PD-L1 expression status of c-CCCs was investigated using two different PD-L1 clones (SP263 and 22C3). PD-L1 expression scores were almost completely similar between the two clones. According to the results of a meta-analysis of the diagnostic accuracy of the PD-L1 IHC assays, the diagnostic sensitivity of the 22C3 was higher than the SP263 assay(34). The fact that our results between the two clones were almost similar for both TPS and CPS scores may be due to the small number of patients.

PD-L1 expression in SCC and ECA has been previously reported(35,36). According to their results, SCC had significantly higher PD-L1 expression positivity in tumor cells than ECAs (5%  cut-off). Omenai et al.(35) recently published the PD-L1 expression profiles of 183 cervical cancer patients, irrespective of their histological type. According to their results, PD-L1 positivity was seen in 57.4% of the cases (58.7% in SCCs and 50% in ECAs). Song et al.(9) shared their results on PD-L1 expression and immune stormal features in HPVI cervical adenocarcinomas. According to their results, PD-L1 expression was seen in 58.3% (7 of 12) of c-CCCs. Moreover, they also found that PD-L1-positive cases (CPS ≥1) showed worse PFS and OS than PD-L1-negative cases. However, data in the literature regarding the PD-L1 expression status of CCCs, the prognostic effect of this expression, and their response to treatment are quite limited. Zong et al.(10) investigated the expression of different immune checkpoint proteins in c-CCCs. They found that 22% of cases had PD-L1-positive tumor cells (CPS ≥1).

Diffuse PD-L1 expression was observed in 18% of our cohort, and the mean PD-L1 expression rate in our c-CCC series was 16%.

Even not statistically significant high TIL density was seen more frequently in PD-L1-positive cases. PD-L1 positivity has been shown to be associated with the number of TILs in many tumor types, including cervical carcinoma(37). As is known, TIL density is likely related to immunotherapy response(38). Similarly, Song et al.(9) showed a significant association between high TIL percentage and CPS or TPS-based PD-L1 expression in their c-CCC cohorts. Additionally, PFS and OS were significantly poorer for PD-L1-positive subjects in their group than for PD-L1-negative cases. Unfortunately, we could not perform a survival analysis due to the short follow-up period in our series and the low number of cases.

Another aim of our study was to investigate the MMR deficiency status in c-CCCs, as the presence of MSI is one of the predictors of anti-PD1/PD-L1 immunotherapy response(39,40). Anti-PD-L1 drugs can enhance survival, particularly in dMMR malignancies(41).

According to our results, 80% of the dMMR cases were also PD-L1 positive and 80% of the dMMR cases had high TILs.  In a recent study involving 39 ECA cases, 2 of which were c-CCC, 15% of the cases were dMMR (all usual type), and 1 dMMR case was PD-L1 positive(42).  Song et al.(9) reported an MMR deficiency status of 16% in their c-CCC cohort (2/12 cases). Both were PD-L1-positive as well. Since the number of studies on the MSI status of c-CCCs is very limited, further studies with larger cohorts will be needed to validate these findings.

Study Limitations

This study has inherent limitations due to the limited sample size owing to the rarity of the disease. As a result, the generalizability of the current findings is limited.


However, our results are valuable in showing that c-CCCs can have PD-L1 expression and MMR deficiency. This shows that c-CCC cases, which are tumor types resistant to conventional chemotherapy, are candidates for immunotherapy similar to other cervical cancer types.


Ethics Committee Approval: This study was approved by the ethics review board of Koç University (approval number: 2023.120.IRB2.028, date: 06.04.2023).

Informed Consent: Informed consent was obtained from all participants.

Peer-review: Internally peer-reviewed.

Authorship Contributions

Conceht: P.B., N.K., Design: P.B., N.K., Data Collection or Processing: P.B., Ö.C.E., Ö.Ö., A.N.H., N.K., Analysis or Interpretation: P.B., N.K., Literature Search: P.B., Ö.C.E., Writing: P.B.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: The authors declared that this study received no financial support.

  1. Stolnicu S, Barsan I, Hoang L, Patel P, Terinte C, Pesci A, et al. International Endocervical Adenocarcinoma Criteria and Classification (IECC): A New Pathogenetic Classification for Invasive Adenocarcinomas of the Endocervix. Am J Surg Pathol 2018;42:214-26.
  2. Lee Y, Bae H, Kim HS. Endocervical Adenocarcinoma: Comprehensive Histological Review and Re-classification of 123 Consecutive Cases According to the Updated World Health Organization Classification of Female Genital Tumors. Anticancer Res 2022;42:4627-39.
  3. Gadducci A, Multinu F, Cosio S, Carinelli S, Ghioni M, Aletti GD. Clear cell carcinoma of the ovary: Epidemiology, pathological and biological features, treatment options and clinical outcomes. Gynecol Oncol 2021;162:741-50.
  4. Mikami Y, Hata S, Melamed J, Moriya T, Manabe T. Basement membrane material in ovarian clear cell carcinoma: correlation with growth pattern and nuclear grade. Int J Gynecol Pathol 1999;18:52-7.
  5. Bulutay P, Haberal N, Özen Ö, Erdem Ö, Zeren EH, Kulac İ, et al. Reproducibility of Morphologic Parameters of the International Endocervical Adenocarcinoma Criteria and Classification System and Correlation With Clinicopathologic Parameters: A Multi-Institutional Study. Int J Gynecol Pathol 2022;41:447-58.
  6. Tewari KS, Sill MW, Penson RT, Huang H, Ramondetta LM, Landrum LM, et al. Bevacizumab for advanced cervical cancer: final overall survival and adverse event analysis of a randomised, controlled, open-label, phase 3 trial (Gynecologic Oncology Group 240). Lancet 2017;390:1654-63.
  7. Colombo N, Dubot C, Lorusso D, Caceres MV, Hasegawa K, Shapira-Frommer R, et al. Pembrolizumab for Persistent, Recurrent, or Metastatic Cervical Cancer. N Engl J Med 2021;385:1856-67.
  8. Da Silva DM, Enserro DM, Mayadev JS, Skeate JG, Matsuo K, Pham HQ, et al. Immune Activation in Patients with Locally Advanced Cervical Cancer Treated with Ipilimumab Following Definitive Chemoradiation (GOG-9929). Clin Cancer Res 2020;26:5621-30.
  9. Song F, Jia M, Yu S, Cao L, Sun PL, Gao H. PD-L1 expression and immune stromal features in HPV-independent cervical adenocarcinoma. Histopathology 2021;79:861-71.
  10. Zong L, Zhang Q, Zhou Y, Kong Y, Yu S, Chen J, et al. Expression and Significance of Immune Checkpoints in Clear Cell Carcinoma of the Uterine Cervix. J Immunol Res 2020;2020:1283632.
  11. Levinson A, Lee AG, Martell HJ, Breese MR, Zaloudek C, Van Ziffle J, et al. Complete Response to PD-1 Inhibition in an Adolescent With Relapsed Clear Cell Adenocarcinoma of the Cervix Predicted by Neoepitope Burden and APOBEC Signature. JCO Precis Oncol 2020;4:PO.20.00132.
  12. Jørgensen JT. An update on companion and complementary diagnostic assays for PD-1/PD-L1 checkpoint inhibitors in NSCLC. Expert Rev Mol Diagn 2021;21:445-54.
  13. Chung HC, Ros W, Delord JP, Perets R, Italiano A, Shapira-Frommer R, et al. Efficacy and Safety of Pembrolizumab in Previously Treated Advanced Cervical Cancer: Results From the Phase II KEYNOTE-158 Study. J Clin Oncol 2019;37:1470-8.
  14. Huang RSP, Haberberger J, Murugesan K, Danziger N, Hiemenz M, Severson E, et al. Clinicopathologic and genomic characterization of PD-L1-positive uterine cervical carcinoma. Mod Pathol 2021;34:1425-33.
  15. de Ruiter EJ, Mulder FJ, Koomen BM, Speel EJ, van den Hout MFCM, de Roest RH, et al. Comparison of three PD-L1 immunohistochemical assays in head and neck squamous cell carcinoma (HNSCC). Mod Pathol 2021;34:1125-32.
  16. Stolnicu S, Barsan I, Hoang L, Patel P, Chiriboga L, Terinte C, et al. Diagnostic Algorithmic Proposal Based on Comprehensive Immunohistochemical Evaluation of 297 Invasive Endocervical Adenocarcinomas. Am J Surg Pathol 2018;42:989-1000.
  17. Alvarado-Cabrero I, Parra-Herran C, Stolnicu S, Roma A, Oliva E, Malpica A. The Silva Pattern-based Classification for HPV-associated Invasive Endocervical Adenocarcinoma and the Distinction Between In Situ and Invasive Adenocarcinoma: Relevant Issues and Recommendations From the International Society of Gynecological Pathologists. Int J Gynecol Pathol 2021;40(Suppl 1):S48-65.
  18. Stolnicu S, Zannoni GF, Soslow RA. Clear Cell Adenocarcinoma In Situ as a Potential Precursor Lesion for Sporadic Invasive Endocervical Clear Cell Adenocarcinoma. Int J Gynecol Pathol 2023;42:217-9.
  19. Binnewies M, Roberts EW, Kersten K, Chan V, Fearon DF, Merad M, et al. Understanding the tumor immune microenvironment (TIME) for effective therapy. Nat Med 2018;24:541-50.
  20. Grau JF, Farinas-Madrid L, Garcia-Duran C, Garcia-Illescas D, Oaknin A. Advances in immunotherapy in cervical cancer. Int J Gynecol Cancer 2023;33:403-13.
  21. Wang T, Lu Z, Zhang X, Hua K. Factors Associated with Patient Survival in Clear Cell Adenocarcinoma of the Cervix: A Single-Center Experience in China. Int J Gen Med 2022;15:4625-34.
  22. Reich O, Tamussino K, Lahousen M, Pickel H, Haas J, Winter R. Clear cell carcinoma of the uterine cervix: pathology and prognosis in surgically treated stage IB-IIB disease in women not exposed in utero to diethylstilbestrol. Gynecol Oncol 2000;76:331-5.
  23. Kasamatsu T, Onda T, Sawada M, Kato T, Ikeda S, Sasajima Y, et al. Radical hysterectomy for FIGO stage I-IIB adenocarcinoma of the uterine cervix. Br J Cancer 2009;100:1400-5.
  24. Liu Z, Li J, Gu H, Tu H, Liu G, Liu J. Clear Cell Adenocarcinoma of Uterine Cervix: A Single Institution Retrospective Experience. Front Oncol 2020;10:532748.
  25. Thomas MB, Wright JD, Leiser AL, Chi DS, Mutch DG, Podratz KC, et al. Clear cell carcinoma of the cervix: a multi-institutional review in the post-DES era. Gynecol Oncol 2008;109:335-9.
  26. Stolnicu S, Karpathiou G, Guerra E, Mateoiu C, Reques A, Garcia A, et al. Clear Cell Carcinoma (CCC) of the Cervix Is a Human Papillomavirus (HPV)-independent Tumor Associated With Poor Outcome: A Comprehensive Analysis of 58 Cases. Am J Surg Pathol 2022;46:765-73. Erratum in: Am J Surg Pathol 2022;46:1317.
  27. Kojima A, Shimada M, Mikami Y, Nagao S, Takeshima N, Sugiyama T, et al. Chemoresistance of Gastric-Type Mucinous Carcinoma of the Uterine Cervix: A Study of the Sankai Gynecology Study Group. Int J Gynecol Cancer 2018;28:99-106.
  28. Abu-Rustum NR, Yashar CM, Bean S, Bradley K, Campos SM, Chon HS, et al. Cervical cancer, Version 1.2020 featured updates to the NCCN guidelines. JNCCN J Natl Compr Cancer Netw 2020;18:660-6.
  29. Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science 2017;357:409-13.
  30. Frenel JS, Le Tourneau C, O’Neil B, Ott PA, Piha-Paul SA, Gomez-Roca C, et al. Safety and Efficacy of Pembrolizumab in Advanced, Programmed Death Ligand 1-Positive Cervical Cancer: Results From the Phase Ib KEYNOTE-028 Trial. J Clin Oncol 2017;35:4035-41.
  31. Li G, Cheng M, Hong K, Jiang Y. Clinical Efficacy and Safety of Immunotherapy Retreatment in Metastatic Cervical Cancer: A Retrospective Study. Onco Targets Ther 2023;16:157-63.
  32. Tewari KS, Monk BJ, Vergote I, Miller A, de Melo AC, Kim HS, et al. Survival with Cemiplimab in Recurrent Cervical Cancer. N Engl J Med 2022;386:544-55.
  33. Rotman J, den Otter LAS, Bleeker MCG, Samuels SS, Heeren AM, Roemer MGM, et al. PD-L1 and PD-L2 Expression in Cervical Cancer: Regulation and Biomarker Potential. Front Immunol 2020;11:596825.
  34. Torlakovic E, Lim HJ, Adam J, Barnes P, Bigras G, Chan AWH, et al. “Interchangeability” of PD-L1 immunohistochemistry assays: a meta-analysis of diagnostic accuracy. Mod Pathol 2020;33:4-17.
  35. Omenai SA, Ajani MA, Okolo CA. Programme death ligand 1 expressions as a surrogate for determining immunotherapy in cervical carcinoma patients. PLoS One 2022;17:e0263615.
  36. Heeren AM, Punt S, Bleeker MC, Gaarenstroom KN, van der Velden J, Kenter GG, et al. Prognostic effect of different PD-L1 expression patterns in squamous cell carcinoma and adenocarcinoma of the cervix. Mod Pathol 2016;29:753-63.
  37. Möller K, Knöll M, Bady E, Schmerder MJ, Rico SD, Kluth M, et al. PD-L1 expression and CD8 positive lymphocytes in human neoplasms: A tissue microarray study on 11,838 tumor samples. Cancer Biomark 2023;36:177-91.
  38. Han R, Zhang Y, Wang T, Xiao H, Luo Z, Shen C, et al. Tumor immune microenvironment predicts the pathologic response of neoadjuvant chemoimmunotherapy in non-small-cell lung cancer. Cancer Sci 2023;114:2569-83.
  39. Deshpande M, Romanski PA, Rosenwaks Z, Gerhardt J. Gynecological Cancers Caused by Deficient Mismatch Repair and Microsatellite Instability. Cancers (Basel) 2020;12:3319.
  40. Zhao P, Li L, Jiang X, Li Q. Mismatch repair deficiency/microsatellite instability-high as a predictor for anti-PD-1/PD-L1 immunotherapy efficacy. J Hematol Oncol 2019;12:54.
  41. Willis BC, Sloan EA, Atkins KA, Stoler MH, Mills AM. Mismatch repair status and PD-L1 expression in clear cell carcinomas of the ovary and endometrium. Mod Pathol 2017;30:1622-32.
  42. Pistolesi S, Fanelli GN, Giudice F, Garbini F, Naccarato AG, Cosio S, et al. Cervical Adenocarcinoma: A Still Under-investigated Malignancy. Anticancer Res 2023;43:53-8.