Abstract
Background.
Uterine clear cell carcinoma is a rare and aggressive subtype of endometrial carcinoma. Prospective clinical trials have not been feasible for this rare tumor, and data regarding the optimal adjuvant treatment regimen for early-stage uterine clear cell carcinomas is limited. Our study’s objective was to determine if adjuvant chemotherapy or radiation therapy improves patients’ outcomes in stage I and II uterine clear cell carcinoma.
Methods.
Patients with stage I and II uterine clear cell carcinoma were identified at a single institution. All cases were reviewed by a gynecologic pathologist. Both pure and mixed non-serous uterine clear cell carcinomas were included. Primary outcomes were recurrence free survival and overall survival.
Results.
A total of 71 patients were identified including 39 (55%) pure and 32 (45%) mixed clear cell carcinoma. Most patients were FIGO stage IA (77.5%). Most patients (n = 58, 82%) received adjuvant therapy, including 43 (61%) receiving chemotherapy, 50 (70%) receiving radiation therapy, and 35 (49%) receiving both. Recurrence free survival was not significantly different among patients receiving no or <6 cycles of chemotherapy versus patients receiving 6 cycles of chemotherapy (p = 0.39). However, median OS was significantly different among patients receiving no or <6 cycles of chemotherapy versus 6 cycles of chemotherapy (p = 0.004). On univariable analysis, 6 cycles of chemotherapy was significantly associated with improved OS (HR 0.1, 95% CI 0.01–0.07). Presence of LVSI, mutated p53, number of pelvic and para-aortic lymph nodes assessed, adjuvant chemotherapy (any number of cycles), and >2 medical co-morbidities were not significant predictors of OS on univariable analysis. On multivariable analysis, 6 cycles of adjuvant chemotherapy remained a significant predictor of improved OS (HR 0.1, 95% CI 0.01–0.8).
Conclusions.
In this study, administration of 6 cycles of chemotherapy appears to significantly improve OS. This finding suggests consideration of 6 cycles of adjuvant chemotherapy in patients with early-stage uterine clear cell carcinoma, however clinical trials are needed to confirm these findings.
Keywords: Uterine malignancy, Uterine clear cell carcinoma, Adjuvant therapy
1. Introduction
Uterine carcinoma is the most common gynecologic malignancy in the United States [1]. Rates of uterine carcinoma are increasing, and a particularly alarming trend has been the increase in the incidence of non-endometrioid, high-risk histologies [1–5]. Uterine clear cell carcinoma (UCCC) is a rare high-risk histologic subtype of endometrial carcinoma (EC) which is associated with higher stage at diagnosis, relative chemoresistance, increased risk of recurrence, and poorer survival when compared to endometrioid adenocarcinomas [6–9]. Due to the rarity of UCCC, prospective clinical trials in the affected patient population are not feasible. Treatment decisions have therefore been guided by subanalyses of clinical trials with a small proportion of UCCC patients or retrospective studies.
Data is particularly limited for UCCC patients with stage I and II disease. Few studies have attempted to address whether a benefit exists for adjuvant radiation therapy and/or chemotherapy in early-stage UCCC, and the results from these studies are mixed [10–16]. Given costs and toxicity associated with chemotherapy and radiation therapy (RT), it is imperative to continue investigation into the role of adjuvant therapy for this patient population.
Another challenge in the study and treatment of UCCC is confirmation of the histologic diagnosis. Accurate pathologic analyses of high-grade ECs can be challenging due to significant overlap in the morphologic and immunohistochemical features [17]. Additionally, previous research suggests that the clinical behavior of mixed clear cell ECs are driven by the more aggressive clear cell component, including those with a small fraction of UCCC [18]. Therefore, review of high-grade EC cases with a specialist gynecologic pathologist is critical in ensuring the diagnosis for both prognostic counseling and treatment decision-making purposes.
The goal of the present study was to determine whether adjuvant chemotherapy, radiation therapy, or both were associated with differences in recurrence-free survival (RFS) and overall survival (OS) in stage I and II UCCC confirmed by specialist gynecologic pathology review.
2. Methods
2.1. Data collection
Institutional Review Board (IRB) approval was obtained for the performance of this study (STUDY19110190). All patients with stage I and II UCCC treated at Magee-Womens’ Hospital of the University of Pittsburgh Medical Center (UPMC) between January 2006 and December 2019 were identified using an institutional pathology database. Data collection was performed using an electronic medical record and included demographics, clinical and pathologic characteristics, adjuvant treatment type, and outcomes. Only patients who had undergone total hysterectomy and bilateral salpingo-oophorectomy were considered for inclusion. For consistency, the 2009 International Federation of Gynecology and Obstetrics (FIGO) staging for endometrial carcinoma was used to define stage, including staging reassignment for cases prior to 2009 [19]. All cases were reviewed by a gynecologic pathologist for histologic confirmation. Given previous literature findings that mixed clear cell histologies have similar clinical behavior to pure UCCC, the decision was made to include both pure UCCC cases and cases of mixed clear cell histology in this study [18].
2.2. Definitions
Chemotherapy regimens that included carboplatin plus a taxane-derivative (e.g. paclitaxel or docetaxel) were included in a single category for descriptive and statistical purposes. Recurrences were categorized by the most distant site of recurrent disease. Cause of death was defined as 1) uterine cancer-related, 2) a known cause not related to the patient’s uterine cancer diagnosis, or 3) unknown.
2.3. Statistical analysis
Primary outcomes included RFS and OS. Chi-square or Fisher exact tests were used to compare the balance of prognostic factors across groups, including by number of chemotherapy cycles, receipt of adjuvant RT, and receipt of both chemotherapy and RT (Supplementary Tables 3–5). Kaplan–Meier failure functions were estimated by pathologic and clinical characteristics, and tested by log-rank tests. For each patient characteristic, a Cox proportional hazards model was fitted, with follow-up days in study as the timescale. Measures of associations were reported as hazard ratios (HRs) with 95% confidence intervals (CIs). Individual associations of all patient characteristics were first assessed in univariable Cox proportional hazards models. Factors yielding statistically significant univariable associations with outcomes of interest and variables postulated in the literature to be related to prognosis of patients were included in multivariable models to yield adjusted estimates of HRs. For RFS and OS, variables tested in the multivariable analysis included presence of lymphovascular space invasion (LVSI), p53 mutation status, pelvic lymph node dissection, ≥4 pelvic lymph nodes assessed, ≥3 para-aortic lymph nodes assessed, receipt of adjuvant chemotherapy, receipt of 6 cycles of adjuvant chemotherapy, >2 medical co-morbidities, diabetes, hypertension, renal disease, and pulmonary disease. Additionally, age, race, BMI, menopausal status, tobacco use, history of hormone replacement therapy use, history of oral contraceptive use, omental sampling, sentinel lymph node assessment, FIGO stage, type of chemotherapy, receipt of adjuvant RT, and type of RT were tested but did not yield statistically significant HRs in univariable models and did not statistically significantly impact adjusted HRs in multivariable models. Percent of clear cell component was included in the multivariable model as a covariate. Violations of the proportional hazards assumption were explored by testing for a zero slope in the scaled Schoenfeld residuals and also by testing time-interaction covariates for statistical significance of the models. The number of pure clear cell endometrial carcinoma was not large enough as a single subset to support robust statistical analyses. Statistical analyses were planned a priori to include analyses of patients receiving 6 cycles of chemotherapy versus no chemotherapy, and 6 cycles of chemotherapy versus none or <6 cycles of chemotherapy. We also performed analyses excluding the 10 patients receiving <6 cycles of chemotherapy, as this group of patients might be associated with factors confounding their survival outcomes, but were too small as a group for subset statistical analyses controlling for these other confounders with adequate power. As a result, comparisons were limited to patients receiving a full 6 cycles of chemotherapy versus patients receiving <6 cycles or no chemotherapy, and our ability to perform robust multivariable analyses with finely categorized variables or subset analyses to evaluate for other confounding factors that would affect survival was limited. Sensitivity analyses on the number of pelvic lymph nodes assessed and percent of clear cell component were conducted to evaluate for relevant cutoffs for categorizations if needed. All analyses were performed using Stata/SE version 15 (StataCorp, College Station, TX).
3. Results
A total of 71 patients were identified that met the inclusion criteria for the study. Demographic and clinicopathologic characteristics are detailed in Table 1. The median age of the cohort was 68 years (range 40–90 years), and 93% of patients were white. Average BMI was 30 kg/m2 (range 18.4–44), and the mean number of medical comorbidities was 2.1 (range 0–10). Of the 71 patients, 39 (54.9%) had pure UCCC, while 32 (45.1%) had mixed clear cell histology. Among patients with mixed histology, 94% (n = 30) had mixed endometrioid and clear cell histology, while 2 patients (6%) had mixed endometrioid, clear cell, and mucinous histology. Mixed histology tumors had a range of 2–90% clear cell carcinoma component.
Table 1.
Demographic and clinicopathologic characteristics of study cohort.*
| N (%) | ||
|---|---|---|
|
| ||
| Age at diagnosis (years) | <60 | 16 (23) |
| 60–75 | 40 (56) | |
| >75 | 15 (21) | |
| Race | White | 66 (93) |
| Black | 3 (4) | |
| Other | 2 (3) | |
| BMI (kg/m2) | <25 | 21 (30) |
| 25–30 | 16 (23) | |
| >30 | 32 (46) | |
| Menopausal status | Premenopausal | 1 (1) |
| Postmenopausal | 69 (97) | |
| Unknown | 1 (1) | |
| Tobacco use | Never | 54 (76) |
| Current | 4 (6) | |
| Former | 10 (14) | |
| History of HRT use | Yes | 6 (8) |
| History of OCP use | Yes | 7 (10) |
| Medical comorbidities | ||
| >2 comorbidities | 24 (34) | |
| History of diabetes | 14 (20) | |
| History of hypertension | 38 (54) | |
| History of renal disease | 3 (4) | |
| History of pulmonary disease | 10 (14) | |
| Type of surgery | ||
| Open | 22 (31) | |
| Laparoscopic | 46 (65) | |
| LAVH/TVH | 3 (4) | |
| Histology | ||
| Pure clear cell | 39 (55) | |
| Mixed clear cell | 32 (45) | |
| p53 mutation | ||
| Yes | 16 (23) | |
| No | 16 (23) | |
| Not assessed | 39 (55) | |
| LVSI | Positive | 22 (31) |
| Peritoneal cytology | ||
| Negative | 58 (87) | |
| Atypical | 2 (3) | |
| Positive | 7 (10.0) | |
| FIGO stage | ||
| IA | 54 (76) | |
| IB | 12 (17) | |
| II | 5 (7) | |
| Lymph node assessment | 63 (88.7) | |
| Sentinel LN | 22 (31) | |
| Pelvic LN | 63 (89) | |
| Assessed ≥ 4 pelvic LN | 51 (72) | |
| Assessed ≥ 3 para-aortic LN | 16 (23) | |
| Adjuvant therapy | Any | 58 (82) |
| Chemotherapy | 43 (61) | |
| Radiation therapy | 50 (70) | |
| Both | 35 (49) | |
| Type of chemotherapy | ||
| Carboplatin/taxane | 41 (95) | |
| Carboplatin alone | 1 (2) | |
| Cisplatin (chemosensitizing) | 1 (2) | |
| Type of radiation therapy | ||
| VBT | 32 (64) | |
| EBRT | 2 (4) | |
| VBT and EBRT | 16 (32) | |
| Recurrence | ||
| Local | 1 (1) | |
| Regional | 3 (4) | |
| Distant | 3 (4) | |
| Cause of death | ||
| Uterine cancer | 4 (6) | |
| Other | 3 (4) | |
| Unknown | 2 (3) | |
Categories may not sum to 100% due to missing data.
The majority of patients underwent laparoscopic staging (65%). Approximately 89% of the cohort had a lymph node assessment, with a mean of 8.3 (range 0–26) pelvic lymph nodes and a mean of 1.5 (range 0–11) para-aortic lymph nodes. Omental sampling was performed in 49 patients (69%). Median tumor size was 28.5 mm (interquartile range [IQR] = 35 mm). The majority of patients were FIGO stage IA (76%) on final pathologic staging.
Fifty-eight (82%) received adjuvant therapy, with 43 (61%) receiving chemotherapy, 50 (70%) receiving RT, and 35 (49%) receiving both. The most common chemotherapy regimen was carboplatin and taxane (n = 41). One patient received cisplatin for chemo-sensitization with radiation therapy, and one patient received carboplatin alone due to chemotherapy reactions to both paclitaxel and docetaxel. Of patients receiving chemotherapy, 10 patients (14%) received <6 cycles, while 32 patients (45%) received a full course of 6 cycles. One patient in the cohort who received chemotherapy did not have information available on the number of cycles received. The most common reasons for receiving <6 cycles of chemotherapy were inability to continue treatment due to toxicities or provider recommendation for <6 cycles. The group of patients receiving <6 cycles of chemotherapy did not yield statistically significant association with OS or RFS compared to the other groups in this cohort, either due to lack of power given the small sample size and/or lack of true association in the population (Supplemental Tables 1 and 2).
The most commonly administered RT was vaginal brachytherapy (VBT) (n = 32), followed by combined external beam radiation therapy (EBRT) and VBT (n = 16), and EBRT alone (n = 2). Recurrences were rare among this cohort, and only 1 patient experienced a local recurrence. Regional and distant recurrences were each noted in 3 patients, respectively. Deaths were also relatively rare among the cohort (n = 9), and were most likely to be uterine cancer-related (n = 4, 6%).
Chi-square or Fisher exact tests were used to compare the balance of selected demographic and clinicopathologic characteristics between groups by cycles of chemotherapy, receipt of adjuvant RT, and the combined variable of chemotherapy and RT (Supplemental Tables 3–5). Among patients receiving none, <6 cycles, and 6 cycles of chemotherapy, the distribution of race, pelvic lymph node assessment, ≥4 pelvic lymph nodes assessed, ≥3 para-aortic lymph nodes assessed, and receipt of adjuvant RT was statistically significantly different between groups (Supplemental Table 3). Among patients who did and did not receive adjuvant RT, age, history of hypertension, presence of LVSI, sentinel lymph node assessment, pelvic lymph node assessment, receipt of adjuvant chemotherapy, and number of cycles of chemotherapy was statistically significantly different between groups (Supplemental Table 4). Among patients receiving both chemotherapy and RT, chemotherapy alone, and RT alone, the distribution of age, presence of LVSI, sentinel lymph node assessment, pelvic lymph node assessment, ≥4 pelvic lymph nodes assessed, and ≥3 para-aortic lymph nodes assessed was statistically significantly different between groups (Supplemental Table 5).
Median follow-up was 35.4 months (IQR = 34.9 mo). For the entire cohort, 5-year RFS was 88% (95% CI 0.74–0.94), and 5-year OS was 85% (95% CI 0.70–0.93). Five-year RFS and OS were similar among patients who did and did not receive adjuvant RT. For those who received adjuvant RT, 5-year RFS was 85% (95% CI 0.71–0.93) and OS was 87% (95% CI 0.68–0.95). Patients who did not receive adjuvant RT had a 5-year RFS of 88% (95% CI 0.39–0.98) and OS of 88% (0.62–0.97). Five-year RFS was 94% (95% CI 0.65–0.99) for patients who did not receive chemotherapy and was 82% (95% CI 0.64–0.92) for patients who did receive adjuvant chemotherapy. Five-year OS was 90% (0.75–0.96) among patients receiving adjuvant chemotherapy, while it was 79% (0.32–0.95) among those who did not receive adjuvant chemotherapy. Patients who received 6 cycles of chemotherapy had a 5-year RFS of 88% (95% CI 0.66–0.96) and OS of 100% compared to a 5-year RFS of 53% (95% CI 0.13–0.83) and OS of 46% (95% CI 0.11–0.76) for those receiving <6 cycles of chemotherapy.
Median RFS for the cohort was not able to be estimated as more than half of the cohort was censored (i.e. did not have a recurrence observed during the follow-up period). There was no difference in RFS among patients receiving no chemotherapy or <6 cycles of chemotherapy versus patients receiving 6 cycles of chemotherapy (p = 0.39) (Fig. 1; see Table 2 for HRs of features associated with RFS). Median OS for the entire cohort was also unable to be estimated as more than half of the cohort did not have an event observed during the follow-up period. Notably, median OS was significantly different among patients receiving no chemotherapy or <6 cycles of chemotherapy versus those receiving 6 cycles of chemotherapy (p = 0.004) (Fig. 2). On univariable analysis, receiving 6 cycles of chemotherapy was significantly associated with improved OS (HR 0.1, 95% CI 0.01–0.7), while renal disease was significantly associated with decreased OS (HR 11.6, 95% CI 2.1–63.4) (Table 3). On multivariable analysis, receiving 6 cycles of adjuvant chemotherapy remained a significant predictor of improved OS (HR 0.1, 95% CI 0.01–0.8) (Table 3). The variable of RT, the combined variable of RT and chemotherapy, and the interaction term of RT and chemotherapy were tested in the survival analyses, and no statistically significant association was seen with any survival outcome.
Fig. 1.
Recurrence-free survival among patients receiving <6 cycles of chemotherapy versus those receiving 6 cycles of chemotherapy.
Table 2.
Hazard ratios of features associated with recurrence-free survival.
| Select Characteristics |
n = 71 n (%)a |
Unadjusted HR (95% CI)**c | Adjusted HR (95% CI)*** | |
|---|---|---|---|---|
|
| ||||
| LVSI | No | 49 (69) | Ref | Ref |
| Yes | 22 (31) | 1.7 (0.4–7.5) | 1.9 (0.4–8.5) | |
| p53 mutationb | No | 16 (23) | Ref | * |
| Yes | 16 (23) | 0.8 (0.3–2.1) | ||
| Assessed ≥4 pelvic lymph nodes | No | 20 (28) | Ref | * |
| Yes | 51 (72) | 1.7 (0.2–14) | ||
| Assessed ≥3 Para-aortic lymph nodes | No | 55 (77) | Ref | * |
| Yes | 16 (23) | 0.4 (0.05–3.2) | ||
| Received adjuvant chemotherapy | No | 28 (39) | Ref | * |
| Yes | 43 (61) | 2.6 (0.3–21.8) | ||
| Received 6 cycles of adjuvant chemotherapy | No chemotherapy | 28 (40) | Ref | * |
| Chemotherapy (<6 cycles) | 10 (14) | 8.5 (0.9–82.5) | ||
| Chemotherapy (6 cycles) | 32 (45) | 1.6 (0.2–15.1) | ||
| Received 6 cycles of adjuvant chemotherapy | No chemotherapy or < 6 cycles | 38 (54) | Ref | Ref |
| 6 cycles of chemotherapy | 32 (45) | 0.5 (0.1–2.4) | 0.4 (0.1–2.0) | |
| >2 co-morbidities | No | 47 (66) | Ref | Ref |
| Yes | 24 (34) | 0.4 (0.05–3.4) | 0.3 (0.05–2.7) | |
| Diabetes | No | 56 (79) | Ref | * |
| Yes | 14 (20) | 2.5 (0.5–13.0) | ||
| Hypertension | No | 32 (46) | Ref | * |
| Yes | 38 (54) | 0.3 (0.1–1.7) | ||
| Pulmonary disease | No | 60 (85) | Ref | * |
| Yes | 10 (14) | 2.5 (0.5–12.8) | ||
Not a significant predictor after controlling for other variables.
Univariable model on age, T stage, peritoneal cytology, history of renal disease do not converge well or do not yield statistically significant associations.
Multivariable model includes percent of clear cell component as covariate.
Percentages may not add up to 100% due to missing data.
Missing data or mutation status not assessed in a subset of cases.
Additionally, age, race, BMI, menopausal status, tobacco use, history of hormone replacement therapy use, history of oral contraceptive use, omental sampling, sentinel lymph node assessment, FIGO stage, type of chemotherapy, receipt of adjuvant radiation therapy, and type of radiation therapy were tested but did not yield statistically significant HRs in univariable models and did not statistically significantly impact adjusted HRs in multivariable models.
Fig. 2.
Overall survival among patients receiving <6 cycles of chemotherapy versus those receiving 6 cycles of chemotherapy.
Table 3.
Hazard ratios of features associated with overall survival.
| Select Characteristics | n = 71 n (%)a |
Unadjusted HR (95% CI)**c | Adjusted HR (95% CI)*** | |
|---|---|---|---|---|
|
| ||||
| LVSI | No | 49 (69) | Ref | Ref |
| Yes | 22 (31) | 0.4 (0.05–3.2) | 0.6 (0.1–5.5) | |
| p53 mutationb | No | 16 (23) | Ref | |
| Yes | 16 (23) | 1.2 (0.4–2.8) | ||
| Pelvic lymph nodes assessed | No | 8 (11) | Ref | |
| Yes | 63 (89) | 0.3 (0.06–1.1) | ||
| Assessed > 4 pelvic lymph nodes | No | 20 (28) | Ref | * |
| Yes | 51 (72) | 0.4 (0.1–1.8) | ||
| Assessed > 3 Para-aortic lymph nodes | No | 55 (77) | Ref | * |
| Yes | 16 (23) | 0.2 (0.03–2.1) | ||
| Received adjuvant chemotherapy | No | 28 (39) | Ref | |
| Yes | 43 (61) | 0.6 (0.1–2.4) | ||
| Received 6 cycles of adjuvant chemotherapy | No chemotherapy | 28 (40) | Ref | |
| Chemotherapy (<6 cycles) | 10 (14) | 2.7 (0.6–12.8) | ||
| Chemotherapy (6 cycles) | 32 (45) | 0.1 (0.01–1.3) | ||
| Received 6 cycles of adjuvant chemotherapy | No chemotherapy or <6 cycles | 38 (54) | Ref | Ref |
| 6 cycles of chemotherapy | 32 (45) | 0.1 (0.01–0.7) | 0.1 (0.01–0.8) | |
| >2 co-morbidities | No | 47 (66) | Ref | Ref |
| Yes | 24 (34) | 4.3 (1.0–18.3) | 3.9 (0.8–19.2) | |
| Diabetes | No | 56 (79) | Ref | * |
| Yes | 14 (20) | 1.9 (0.4–9.9) | ||
| Hypertension | No | 32 (46) | Ref | * |
| Yes | 38 (54) | 1.2 (0.3–5.1) | ||
| Renal disease | No | 67 (96) | Ref | * |
| Yes | 3 (4) | 11.6 (2.1–63.4) | ||
| Pulmonary disease | No | 60 (86) | Ref | * |
| Yes | 10 (14) | 3.1 (0.7–13.3) | ||
Not a significant predictor after controlling for other variables.
Univariable model on age, T stage and peritoneal cytology do not converge well or do not yield statistically significant associations.
Multivariable model includes percent of clear cell component as covariate. Received >6 cycles of chemotherapy yields statistically significant HR whether adjusting for percent of clear cell component as continuous or categorical variables.
Percentages may not add up to 100% due to missing data.
Missing data or mutation status not assessed in a subset of cases.
Additionally, age, race, BMI, menopausal status, tobacco use, history of hormone replacement therapy use, history of oral contraceptive use, omental sampling, sentinel lymph node assessment, FIGO stage, type of chemotherapy, receipt of adjuvant radiation therapy, and type of radiation therapy were tested but did not yield statistically significant HRs in univariable models and did not statistically significantly impact adjusted HRs in multivariable models.
Additionally, statistical analyses were performed which excluded the 10 patients receiving <6 cycles of chemotherapy. In the cohort of patients excluding those receiving <6 cycles of chemotherapy (n = 61), patients treated with 6 cycles of chemotherapy had better overall survival (OS) when compared to those who did not receive chemotherapy (log rank p = 0.01 (Supplemental Fig. 1). Receiving 6 cycles of chemotherapy predicted improved OS in the univariable model (HR: 0.01, 95% CI: 0.01–0.9) and when controlling for percent of clear cell component in the tumor (aHR: 0.05 (95% CI: 0.002–0.9) in the multivariable model (Supplemental Table 6). While the HR did not reach statistical significance when including other potential confounding variables in the multivariable model, the relative small sample size and low number with outcome of interest limited the extent of confounder evaluation with adequate power in this cohort (Supplemental Table 6). Receiving 6 cycles of chemotherapy was not found to be a statistically significant predictor of recurrence-free survival in this cohort (Supplemental Table 7 and Supplemental Fig. 2).
4. Discussion
UCCC is a rare and aggressive histologic subtype of EC. Data to guide treatment decisions for this patient population are lacking due to the inability to perform prospective clinical trials and the limited number of patients available for subgroup analyses. In particular, there are few studies that have attempted to assess the role of adjuvant therapy for patients with early-stage UCCC, leading to decisions based on individual provider experience and patient preference. Therefore, the objective of our study was to investigate the role of adjuvant therapy for patients with stage I and II UCCC after confirmation with a gynecologic pathologist.
In the current study, 61% of patients received adjuvant chemotherapy, and the most frequent chemotherapy regimen was a combination of carboplatin and taxane therapy. Interestingly, there was no difference in RFS between patients who received 0–5 cycles of chemotherapy versus patients who received 6 cycles. However, OS was statistically significantly different between groups, with improved OS in those patients who received 6 cycles of chemotherapy. We suspect that the discrepancy between RFS and OS can be explained by the limited sample size and the small number of recurrences observed in this study. We would expect differences in RFS with regard to adjuvant chemotherapy to become more apparent in a larger patient population. We also acknowledge that while patients receiving no chemotherapy may differ in fundamental ways from patients receiving <6 cycles of chemotherapy, the ability to perform robust multivariable analyses between these groups was limited by the sample size of patients in our cohort. Nonetheless, it is impressive that such notable differences in OS were observed between treatment groups despite the constraints of a small patient cohort.
Current National Comprehensive Cancer Network (NCCN) recommendations for stage IA UCCC include systemic chemotherapy and VBT as the preferred adjuvant treatment regimen. For patients with stage IB and stage II disease, recommendations include systemic chemotherapy with or without EBRT and VBT [20]. Thomas et al. performed a multi-institutional review of outcomes for UCCC patients after surgical staging. Their study included 49 patients with stage I-II disease, of whom 10 developed recurrent disease. Among the 22 patients who had definitive staging with lymphadenectomy, half were observed and half received adjuvant therapy with RT, chemotherapy or both. Only 1 recurrence was noted in this group, which was located at the vaginal cuff. However, comparisons between adjuvant therapy modalities in stage I and II patients were limited by the small number of patients receiving each type of therapy. When stage I, II, and III patients were combined, adjuvant RT was an independent predictor of OS and PFS, suggesting that adjuvant RT could be considered for treatment of early-stage UCCC [10].
GOG 249 compared RT versus VBT plus chemotherapy in patients with high-intermediate and high-risk early-stage EC. Unfortunately, only 28 patients with UCCC were enrolled in the trial. Subgroup analyses from GOG 249 revealed no difference in outcomes between treatment arms for serous and UCCC patients, although the overall trial results failed to show improvement with vaginal brachytherapy and chemotherapy compared to RT alone [11]. Another study by Zhang et al. compared outcomes between early-stage patients with UCCC and UPSC. While UCCC patients were significantly less likely to receive adjuvant chemotherapy, outcomes were similar to patients with UPSC with regard to recurrence rate, OS, and site of recurrence, suggesting that chemotherapy may be omitted in early-stage UCCC. However, it is important to recognize that the conclusions were again limited by a small number of patients, with only 17 UCCC patients included, of whom only 6 received adjuvant chemotherapy [12]. A large NCDB analysis of adjuvant therapy for 2516 patients early-stage UCCC additionally failed to find evidence of statistically significant improvement in OS with adjuvant chemotherapy or RT. The authors did note a trend towards improvement in OS for women undergoing hysterectomy, lymphadenectomy and chemotherapy when compared to patients without lymphadenectomy, chemotherapy or both regardless of RT. [13].
In addition to the logistical and practical considerations involved in the study of adjuvant treatment of UCCC, obtaining an accurate pathologic diagnosis can present challenges. Grade 3 endometrioid, uterine papillary serous carcinoma, and UCCC share a large number of morphologic and immunohistochemical features that can make it difficult to differentiate between subtypes. One study by Fadare et al. demonstrated that the diagnosis of UCCC can be variable even among gynecologic pathologists secondary to the high degree of morphologic variety between tumors [21]. Additionally, tumors with true mixed histology add another layer of complexity to the diagnostic evaluation. Prior research has demonstrated that tumors with mixed histology display clinical behavior that mimics the more aggressive histologic component, making accurate diagnosis even more imperative [18]. One of the primary strengths of this study was the pathologic review of each individual patient’s tumor histology by a gynecologic pathologist to confirm the diagnosis of pure or mixed UCCC. Larger studies of UCCC that have failed to find differences between adjuvant treatment groups may have been limited by lack of diagnostic accuracy, as pathology review is time-intensive and requires access to patient’s tumor samples.
It is important to acknowledge that the primary objective of this study was to determine whether adjuvant therapy was associated with improvements in RFS and OS, and that the significant improvement seen in OS with a full 6 cycles of chemotherapy was detected on exploratory analysis. Another important limitation of this study was the lack of racial and ethnic diversity in the patient population. Clear racial disparities exist for black patients compared to white patients, including a higher incidence of aggressive histologic subtypes, worse stage at presentation, and worse survival outcomes even when controlling for disease-related factors. The lack of diversity in our cohort can be attributed to the limitations of a single institution study, and it is possible that differences in prognostic variables and outcomes could be seen in a more well-represented patient population [22]. Additional limitations of this study include those inherent to a retrospective cohort study, such as missing data, changes in practice patterns over time, and differences among providers in management and treatment decisions. Lastly, this study was limited by constraints on sample size due to the rarity of UCCC. However, we feel that the confirmed pathologic diagnosis for each patient and the information available on adjuvant treatment and survival make this study an important contribution to the literature for this uncommon disease. While additional studies are needed, our results support the role of a full course of adjuvant chemotherapy in patients with stage I and II UCCC.
Supplementary Material
HIGHLIGHTS.
Pathologic confirmation of pure or mixed uterine clear cell carcinoma was performed for each patient.
Receipt of adjuvant therapy was not associated with improvements in RFS.
Receipt of a full 6 cycles of adjuvant chemotherapy was associated with improved OS in this cohort.
Funding
The authors did not receive financial support for this project.
Footnotes
Declaration of Competing Interest
Dr. Beriwal is a consultant for Elsevier Pathway and a member of the Data Safety Monitoring Board for Xoft. Dr. Beriwal was a faculty member in the Department of Radiation Oncology at the University of Pittsburgh Medical Center at the time that this project was conceived and conducted; he is now employed at Varian Medical Systems and Allegheny Health Network. Dr. Buckanovich is a cofounder of Tradewinds Bioscience. Dr. Bhargava has served as a consultant for Agilent Technologies, Inc. and ImmunoGen. Otherwise, the authors have no relevant conflicts of interest to disclose.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ygyno.2022.12.024.
References
- [1].Siegel RL, Miller KD, Jemal A, Cancer statistics, 2020, CA Cancer J Clin 70 (2020) 7–30, 10.3322/caac.21590. [DOI] [PubMed] [Google Scholar]
- [2].Faber MT, Frederiksen K, Jensen A, Aarslev PB, Kjaer SK, Time trends in the incidence of hysterectomy-corrected overall, type 1 and type 2 endometrial cancer in Denmark 1978–2014, Gynecol Oncol 146 (2017) 359–367, 10.1016/j.ygyno.2017.05.015. [DOI] [PubMed] [Google Scholar]
- [3].Doll KM, Winn AN, Assessing endometrial cancer risk among US women: long-term trends using hysterectomy-adjusted analysis, Am J Obstet Gynecol 221 (2019) 318.e1–318.e9, 10.1016/j.ajog.2019.05.024. [DOI] [PubMed] [Google Scholar]
- [4].Lortet-Tieulent J, Ferlay J, Bray F, Jemal A, International patterns and trends in endometrial cancer incidence, 1978–2013, J Natl Cancer Inst 110 (2018) 354–361, 10.1093/jnci/djx214. [DOI] [PubMed] [Google Scholar]
- [5].Clarke MA, Devesa SS, Harvey SV, Wentzensen N, Hysterectomy-corrected uterine corpus cancer incidence trends and differences in relative survival reveal racial disparities and rising rates of nonendometrioid cancers, J Clin Oncol 37 (2019) 1895–1908, 10.1200/JCO.19.00151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Cirisano FD, Robboy SJ, Dodge RK, Bentley RC, Krigman HR, Synan IS, et al. , The outcome of stage I-II clinically and surgically staged papillary serous and clear cell endometrial cancers when compared with endometrioid carcinoma, Gynecol Oncol 77 (2000) 55–65, 10.1006/gyno.2000.5737. [DOI] [PubMed] [Google Scholar]
- [7].Hamilton CA, Cheung MK, Osann K, Chen L, Teng NN, Longacre TA, et al. , Uterine papillary serous and clear cell carcinomas predict for poorer survival compared to grade 3 endometrioid corpus cancers, Br J Cancer 94 (2006) 642–646, 10.1038/sj.bjc.6603012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Altman AD, Ferguson SE, Atenafu EG, Köbel M, McAlpine JN, Panzarella T, et al. , Canadian high risk endometrial cancer (CHREC) consortium: analyzing the clinical behavior of high risk endometrial cancers, Gynecol Oncol 139 (2015) 268–274, 10.1016/j.ygyno.2015.09.001. [DOI] [PubMed] [Google Scholar]
- [9].McMeekin DS, Filiaci VL, Thigpen JT, Gallion HH, Fleming GF, Rodgers WH, et al. , The relationship between histology and outcome in advanced and recurrent endometrial cancer patients participating in first-line chemotherapy trials: a Gynecologic Oncology Group study, Gynecol Oncol 106 (2007) 16–22, 10.1016/j.ygyno.2007.04.032. [DOI] [PubMed] [Google Scholar]
- [10].Thomas M, Mariani A, Wright JD, Madarek EOS, Powell MA, Mutch DG, et al. , Surgical management and adjuvant therapy for patients with uterine clear cell carcinoma: a multi-institutional review, Gynecol Oncol 108 (2008) 293–297, 10.1016/j.ygyno.2007.11.008. [DOI] [PubMed] [Google Scholar]
- [11].Randall ME, Filiaci V, McMeekin DS, von Gruenigen V, Huang H, Yashar CM, et al. , Phase III trial: adjuvant pelvic radiation therapy versus vaginal brachytherapy plus paclitaxel/carboplatin in high-intermediate and high-risk early stage endometrial cancer, J Clin Oncol 37 (2019) 1810–1818, 10.1200/JCO.18.01575 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Zhang M, Yang TJ, Desai NB, DeLair D, Kollmeier MA, Makker V, et al. , Comparison of outcomes in early-stage uterine clear cell carcinoma and serous carcinoma, Brachytherapy 18 (2019) 38–43, 10.1016/j.brachy.2018.08.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Hong JC, Foote J, Broadwater G, Gaillard S, Havrilesky LJ, Chino JP, Impact of chemotherapy and radiotherapy on management of early stage clear cell and papillary serous carcinoma of the uterus, Int J Gynecol Cancer 27 (2017) 720–729, 10.1097/IGC.0000000000000926. [DOI] [PubMed] [Google Scholar]
- [14].Townamchai K, Berkowitz R, Bhagwat M, Damato AL, Friesen S, Lee LJ, et al. , Vaginal brachytherapy for early stage uterine papillary serous and clear cell endometrial cancer, Gynecol Oncol 129 (2013) 18–21, 10.1016/j.ygyno.2012.12.026. [DOI] [PubMed] [Google Scholar]
- [15].Barney BM, Petersen IA, Mariani A, Dowdy SC, Bakkum-Gamez JN, Haddock MG, The role of vaginal brachytherapy in the treatment of surgical stage I papillary serous or clear cell endometrial cancer, Int J Radiat Oncol Biol Phys 85 (2013) 109–115, 10.1016/j.ijrobp.2012.03.011. [DOI] [PubMed] [Google Scholar]
- [16].Shinde A, Li R, Amini A, Chen Y-J, Cristea M, Dellinger T, et al. , Improved survival with adjuvant brachytherapy in stage IA endometrial cancer of unfavorable histology, Gynecol Oncol 151 (2018) 82–90, 10.1016/j.ygyno.2018.08.028. [DOI] [PubMed] [Google Scholar]
- [17].Murali R, Davidson B, Fadare O, Carlson JA, Crum CP, Gilks CB, et al. , High-grade endometrial carcinomas: morphologic and immunohistochemical features, diagnostic challenges and recommendations, Int J Gynecol Pathol 38 (Suppl. 1) (2019) S40–S63, 10.1097/PGP.0000000000000491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Hsu K-F, Chou H-H, Huang C-Y, Fu H-C, Chiang A-J, Tsai H-W, et al. , Prognostic factors and treatment outcomes for patients with surgically staged uterine clear cell carcinoma focusing on the early stage: A Taiwanese Gynecologic Oncology Group study, Gynecol Oncol (2014) 10.1016/j.ygyno.2014.07.005. [DOI] [PubMed] [Google Scholar]
- [19].Pecorelli S, Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium, Int J Gynaecol Obstet 105 (2009) 103–104, 10.1016/j.ijgo.2009.02.012. [DOI] [PubMed] [Google Scholar]
- [20].Uterine Neoplasms, NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) 2021;Version 3, 2021. [Google Scholar]
- [21].Fadare O, Parkash V, Dupont WD, Acs G, Atkins KA, Irving JA, et al. , The diagnosis of endometrial carcinomas with clear cells by gynecologic pathologists: an assessment of interobserver variability and associated morphologic features, Am J Surg Pathol 36 (2012) 1107–1118, 10.1097/PAS.0b013e31825dd4b3. [DOI] [PubMed] [Google Scholar]
- [22].Long B, Liu FW, Bristow RE, Disparities in uterine cancer epidemiology, treatment, and survival among African Americans in the United States, Gynecol Oncol 130 (2013) 652–659, 10.1016/j.ygyno.2013.05.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.