Abstract
Objective.
To examine survival of women who develop metachronous uterine malignancy after definitive pelvic radiotherapy for cervical cancer.
Methods.
This retrospective observational study examined the Surveillance, Epidemiology, End Results Program between 1973 and 2013. Women with cervical cancer who received definitive radiotherapy without hysterectomy were examined for the diagnosis of metachronous uterine malignancy (n = 5277). Survival was compared between metachronous and non-metachronous uterine malignancies according to tumor factors.
Results.
The 10- and 20-year cumulative incidences of metachronous uterine malignancy were 0.6% and 1.2%, respectively. When compared to non-metachronous uterine malignancy, metachronous tumor were more likely to be non-endometrioid and advanced-stage (both, P < 0.001). As a whole cohort, metachronous uterine malignancy was significantly associated with decreased overall survival (OS) compared to non-metachronous tumors (hazard ratio [HR] 4.22, P < 0.001). OS was significantly worse in metachronous compared to non-metachronous malignancies, although the magnitude of statistical significance was greater for endometrioid tumors (HRs for endometrioid versus non-endometrioid: 6.17 versus 1.92). For grade 1–2 endometrial cancer, metachronous cases had significantly decreased OS compared to non-metachronous cases, a larger difference than that seen in higher grade tumors (HRs for grade 1–2 versus 3: 7.79 versus 2.15). Similarly, in early-stage endometrial cancer, metachronous cases had significantly decreased OS, with a greater HR compared to advanced-stage disease (HRs for stage I–II versus III–IV: 5.29 versus 2.29).
Conclusion.
Radiotherapy-associated metachronous uterine malignancy after cervical cancer is rare but commonly presents with aggressive tumor characteristics. The impact on survival is considerably high when metachronous uterine malignancy is endometrioid, low-grade, and early-stage.
Keywords: Cervical cancer, Radiotherapy, Metachronous, Uterine malignancy, Early stage, Low grade
1. Introduction
Carcinoma of the uterine cervix remains the most common gynecologic malignancy worldwide [1], and women in developing countries typically present with locally-advanced tumors while early-stage disease is more common in developed countries. Treatment of women with cervical cancer is largely dependent on cancer stage, and radiation-based therapy plays a pivotal role in the management of women with locally advanced-stage disease or with high- or intermediate-risk early-stage disease [2].
Women with cervical cancer who receive pelvic radiotherapy are at increased risk of developing secondary malignancies [3]. Various studies have shown that secondary primary malignancies can develop in organs within the radiated field after pelvic irradiation, including colo-rectal, urinary bladder, vulvar, vaginal, ovarian and anal malignancies [4-8]. While the incidence is rare, these metachronous tumors commonly develop many years after the primary cancer diagnosis.
Various studies have also shown an increased risk of metachronous uterine malignancies in women with a history of cervical cancer treated with radiotherapy [9,10]. Radiation-associated uterine malignancies commonly present with high-grade tumors and with advanced-stage disease, resulting in poor prognosis [11-13]. Despite the fact that metachronous uterine malignancies after pelvic irradiation usually have high-risk tumor characteristics and a poor outcome, a considerable fraction, >25–30%, of these tumors are early-stage or low-grade [11-13].
To date, tumor characteristics and outcomes related to early-stage or low-grade metachronous uterine malignancy after radiotherapy for cervical cancer have not been examined. The objective of the study was to examine survival of women who develop metachronous uterine malignancy after definitive pelvic radiotherapy for cervical cancer, stratified by tumor factors.
2. Materials and methods
2.1. Study eligibility
This is a retrospective study examining the Surveillance, Epidemiology, and End Results Program. This database is the largest population-based tumor registry in the United States, and it is publicly available and de-identified [14]. This database was launched in 1973 by the National Cancer Institute and currently covers approximately 28% of the US population. Data entry for this database is performed by registered and trained personnel with rigorous quality control [15]. The Institutional Review Board at the University of Southern California exempted this study due to the use of publicly available deidentified data.
The SEER*Stat 8.3.2 (IMS Inc., Calverton, MD, USA) was used to extract the SEER18 cases between 1973 and 2013. First, the dataset was generated for “Cervix Uteri,” limited to malignancy and female sex. Eligible cases were primary cervical cancers who did not undergo hysterectomy and who were treated with whole pelvic radiotherapy. Sarcomas or metastatic tumors to the uterine cervix, performance of hysterectomy or unknown surgical status, or absence of radiotherapy or unknown radiation type were excluded.
To examine metachronous uterine malignancy, the SEER*Stat 8.3.2 (IMS Inc., Calverton, MD, USA) was also utilized to generate a dataset for “Corpus Uteri,” limited to malignancy and female sex during the same time period (1973–2013). The two datasets were then linked, and the study identification numbers assigned by the program were utilized to identify secondary primary cancer cases. Patients identified in both datasets were considered secondary primary cancer cases as described previously [16-18].
2.2. Study definition
The chronologic time sequence was examined between cervical cancer diagnosis and uterine malignancy diagnosis. In this study, when the time interval between the two diagnoses was ≥6 months, the cases were considered to have a metachronous secondary primary cancer based on prior studies [16-19]. Cases with a time interval <6 months, diagnosis antecedent to cervical cancer, and secondary entry were excluded from the study.
2.3. Clinical information
Among the eligible cases for the study, the following information was abstracted from the database: patient baseline demographics, tumor characteristics, treatment types, and survival. Patient demographics were age, year and month at diagnosis, race/ethnicity, marital status, and registration area. Tumor characteristics including cancer stage, histologic subtype, and tumor grade were analyzed. Treatment variables included types of surgery and radiotherapy.
Survival information abstracted from the database included cause-specific survival (CSS), defined as time between diagnosis of the index cancer and death from the index cancer, and overall survival (OS), defined as time between diagnosis of the index cancer and death from any cause. Cases without survival events were censored at the last follow-up. Grouping and categorization of clinico-pathological factors were based on prior studies [20-22].
2.4. Statistical analysis
The primary objective of the study was to examine the cumulative incidence and characteristics of metachronous uterine malignancy in women with cervical cancer who received whole pelvic radiotherapy without hysterectomy. The secondary objective of the study was to examine survival of uterine malignancy according to tumor grade (grade 1–2 versus 3), cancer stage (stage I–II versus III IV), and histology type (endometrioid versus non-endometrioid). Women with metachronous uterine malignancy were compared to those without metachronous tumors.
Normality of continuous variables was assessed by the Kolmogorov–Smirnov test, and differences were assessed by the Student t-test or Mann-Whitney U test as appropriate. Differences in ordinal or categorical variables were assessed by the Fisher exact test or chi-square test as appropriate. The Kaplan-Meier method was utilized to construct cumulative incidence curves or survival curves, and the difference between the curves was assessed by the log-rank test for univariable analysis. A Cox proportional hazard regression model was used to estimate the unadjusted hazard ratio (HR) with 95% confidence interval (CI) for the covariates of interest. Based on the assumption that incidence of metachronous uterine malignancy was <1% with inadequate survival events to conduct a multivariable analysis, multivariable analysis was not pre-planned to avoid overfitting.
All analyses were based on two-tailed hypotheses, and a P < 0.05 was considered statistically significant. Statistical Package for Social Sciences (SPSS, version 24.0, IBM Corp, Armonk, NY, USA) was used for the analysis. The STROBE guidelines were consulted for performance of this observational study.
3. Results
Among 87,151 women with malignancy of the uterine cervix, 15,296 women did not undergo hysterectomy for treatment (Fig. 1). Of those, 5329 women received whole pelvic radiotherapy. After excluding synchronous uterine malignancy and rare histologic types, 5,277 women with primary cervical cancer who received whole pelvic radiotherapy without hysterectomy comprised the study population.
Fig. 1.
Study schema. Abbreviations: WPRT, whole pelvic radiotherapy; and EMCA, endometrial cancer.
The median follow-up time of the entire cervical cancer cohort was 4.8 years (interquartile range 10.8); there were 1,722 and 643 women who had ≥10 and ≥20 years follow-up time, respectively. There were 22 metachronous uterine malignancies that occurred in 5,277 women with cervical cancer. The 5-, 10- and 20-year cumulative incidences of metachronous uterine malignancy were <0.1%, 0.6% and 1.2%, respectively (Fig. 2A). Among the cases of metachronous uterine malignancy, the median time to a diagnosis of uterine malignancy was 8.6 years (interquartile range, 7.8). The majority of the cases (20 out of 22, 90.9%) developed the metachronous uterine malignancy at least 6 years after the cervical cancer diagnosis.
Fig. 2.
Cumulative incidence of uterine malignancy after cervical cancer diagnosis. Log-rank test for P-value. Cumulative incidence of endometrial cancer after cervical cancer diagnosis is shown for (A) all uterine malignancies and (B) endometrioid type endometrial cancer versus non-endometrioid endometrial cancer.
The most common type of metachronous uterine malignancy was endometrial cancer (95.5%) followed by uterine sarcoma (4.5%). Among those with metachronous endometrial cancer, the majority were non-endometrioid (59.1%), followed by endometrioid endometrial cancer (36.4%). The cumulative incidences were similar between women who developed metachronous endometrioid endometrial cancer and those who developed metachronous non-endometrioid endometrial cancer (P = 0.28; Fig. 2B).
Women with cervical cancer who developed metachronous uterine malignancy had similar characteristics to those who did not develop metachronous uterine malignancy with respect to age, race/ethnicity, year of diagnosis, marital status, and registered area (Table 1) (all, P > 0.05). Cervical cancer histologic type, stage, lymph node status, and tumor size were also not associated with development of metachronous uterine malignancy (all, P > 0.05). Cervical cancer tumor grade was significantly associated with development of subsequent metachronous uterine malignancy; women with metachronous uterine malignancy were more likely to have had grade 1 cervical tumors compared to those who did not (33.3% versus 10.5%; P = 0.028).
Table 1.
Patient demographics for cervical cancer.
| Characteristic | Metachronous EMCA (−) |
Metachronous EMCA (+) |
P-value |
|---|---|---|---|
| Number | n = 5255 | n = 22 | |
| Age | 50 (IQR 22) | 52.5 (IQR 29) | 0.70 |
| <50 | 2516 (47.9%) | 9 (40.9%) | |
| ≥50 | 2739 (52.1%) | 13 (59.1%) | |
| Race/ethnicity | 0.53 | ||
| White | 3107 (59.1%) | 10 (45.5%) | |
| Black | 908 (17.3%) | 6 (27.3%) | |
| Hispanic | 819 (15.6%) | 3 (13.6%) | |
| Asian | 324 (6.2%) | 2 (9.1%) | |
| Others | 97 (1.8%) | 1 (4.5%) | |
| Year | 0.07 | ||
| 1973–9 | 878 (16.7%) | 4 (18.2%) | |
| 1980–9 | 738 (14.0%) | 5 (22.7%) | |
| 1990–9 | 1180 (22.5%) | 9 (40.9%) | |
| 2000–9 | 1923 (36.6%) | 4 (18.2%) | |
| 2010–3 | 536 (10.2%) | 0 | |
| Area | 0.60 | ||
| West | 2623 (49.9%) | 13 (59.1%) | |
| Central | 1422 (27.1%) | 4 (18.2%) | |
| East | 1210 (23.0%) | 5 (22.7%) | |
| Marital status | 0.40 | ||
| Married | 2386 (47.1%) | 13 (61.9%) | |
| Single | 983 (19.4%) | 3 (14.3%) | |
| Others | 1700 (33.5%) | 5 (23.8%) | |
| Histology | 0.34 | ||
| Squamous | 4116 (78.3%) | 14 (63.6%) | |
| Adenocarcinoma | 701 (13.3%) | 5 (22.7%) | |
| Adenosquamous | 217 (4.1%) | 2 (9.1%) | |
| Others | 221 (4.2%) | 1 (4.5%) | |
| Grade | 0.028 | ||
| 1 | 373 (10.5%) | 4 (33.3%) | |
| 2 | 1504 (42.3%) | 5 (41.7%) | |
| 3 | 1678 (47.2%) | 3 (25.0%) | |
| Stage | 0.26 | ||
| I | 1408 (34.9%) | 9 (56.3%) | |
| II | 983 (24.4%) | 4 (25.0%) | |
| III | 1032 (25.6%) | 2 (12.5%) | |
| IV | 609 (15.1%) | 1 (6.3%) | |
| Nodal metastasis | 0.77 | ||
| No | 2552 (70.7%) | 10 (76.9%) | |
| Yes | 1057 (29.3%) | 3 (23.1%) | |
| Tumor size | 0.38 | ||
| ≤2.0 cm | 390 (19.6%) | 1 (20.0%) | |
| 2.1–4.0 cm | 525 (26.4%) | 0 | |
| >4.0 cm | 1074 (54.0%) | 4 (80.0%) |
Median (IQR) or number (percent per column) is shown. Mann-Whitney U test, Fisher exact test, or chi-square test for P-values. Significant P-values are emboldened. Abbreviations: EMCA, endometrial cancer; and IQR, interquartile range. Cases with unknown status were excluded from analysis.
The characteristics of the 22 metachronous uterine malignancies were then compared to 245,913 non-metachronous uterine malignancies (Table 2). When compared to women with non-metachronous uterine malignancy, women with metachronous uterine malignancy were less likely to be white (45.5% versus 77.2%, P = 0.003). Metachronous tumors were less likely to be endometrioid type (36.4% versus 72.3%) but more likely to be serous (27.3% versus 5.9%), clear cell (9.1% versus 1.2%), and carcinosarcoma (18.2% versus 4.4%) (Supplemental Table S1, P < 0.001). Metachronous uterine malignancies were also less likely to be stage I (45.5% versus 62.7%) and more likely to present with stage IV disease (31.8% versus 7.5%) compared to non-metachronous malignancies (P = 0.001).
Table 2.
Patient demographics for uterine corpus malignancy.
| Characteristic | Metachronous uterine malignancy (−) |
Metachronous uterine malignancy (+) |
P-value |
|---|---|---|---|
| Number | n = 245,913 | n = 22 | |
| Age | 63 (IQR 17) | 63 (IQR 25) | 0.83 |
| ≥60 | 97,097 (39.5%) | 8 (36.4%) | |
| ≥60 | 148,816 (60.5%) | 14 (63.6%) | |
| Race/ethnicity | 0.003 | ||
| White | 189,952 (77.2%) | 10 (45.5%) | |
| Black | 19,046 (7.7%) | 6 (27.3%) | |
| Hispanic | 19,270 (7.8%) | 3 (13.6%) | |
| Asian | 12,921 (5.3%) | 2 (9.1%) | |
| Others | 4724 (1.9%) | 1 (4.5%) | |
| Year | 0.15 | ||
| 1973–9 | 21,537 (8.8%) | 0 | |
| 1980–9 | 28,493 (11.6%) | 2 (9.1%) | |
| 1990–9 | 41,359 (16.8%) | 7 (31.8%) | |
| 2000–9 | 102,919 (41.9%) | 11 (50.0%) | |
| 2010–3 | 51,605 (21.0%) | 2 (9.1%) | |
| Area | 0.75 | ||
| West | 126,392 (51.4%) | 13 (59.1%) | |
| Central | 57,774 (23.5%) | 4 (18.2%) | |
| East | 61,747 (25.1%) | 5 (22.7%) | |
| Marital status | 0.06 | ||
| Married | 127,279 (54.1%) | 7 (35.0%) | |
| Single | 35,791 (15.2%) | 2 (10.0%) | |
| Others | 72,262 (30.7%) | 11 (55.0%) | |
| Histology* | <0.001 | ||
| Endometrioid | 177,872 (72.3%) | 8 (36.4%) | |
| Non-endometrioid | 57,534 (23.4%) | 13 (59.1%) | |
| Sarcoma | 10,507 (4.3%) | 1 (4.5%) | |
| Grade | <0.001 | ||
| 1 | 85,497 (42.2%) | 1 (5.3%) | |
| 2 | 64,113 (31.6%) | 6 (31.6%) | |
| 3 | 53,203 (26.2%) | 12 (63.2%) | |
| Stage | 0.001 | ||
| I | 154,263 (75.6%) | 10 (50.0%) | |
| II | 10,354 (5.1%) | 1 (5.0%) | |
| III | 20,929 (10.3%) | 2 (10.0%) | |
| IV | 18,454 (9.0%) | 7 (35.0%) | |
| Tumor size | 0.41 | ||
| ≤2.0 cm | 22,338 (21.2%) | 3 (33.3%) | |
| >2.0 cm | 83,087 (78.8%) | 6 (66.7%) | |
| Hysterectomy | 0.72 | ||
| No | 20,146 (9.5%) | 1 (4.8%) | |
| Yes | 191,875 (90.5%) | 20 (95.2%) | |
| Lymphadenectomy | 0.014 | ||
| Not performed | 98,066 (47.2%) | 15 (75.0%) | |
| Performed | 109,883 (52.8%) | 5 (25.0%) | |
| Lymph node number | 12 (IQR 15) | 2.5 (IQR 14) | 0.057 |
| Radiotherapy | 0.035† | ||
| None | 169,755 (70.5%) | 20 (90.9%) | |
| External | 50,481 (21.0%) | 1 (4.5%) | |
| Implant | 13,363 (5.5%) | 1 (4.5%) | |
| Not specified | 7189 (3.0%) | 0 |
Median (IQR) or number (percent per column) is shown. Mann-Whitney U test, Fisher exact test, or chi-square test for P-values. Significant P-values are emboldened.
detail of histology type-specific analysis is shown in Supplemental Table S1.
none versus any radiotherapy. Abbreviations: EMCA, endometrial cancer; and IQR, interquartile range. Cases with unknown status were excluded from analysis.
The vast majority of women with metachronous uterine malignancy underwent hysterectomy (95.2% versus 90.5%, P = 0.72), however, they were less likely to undergo pelvic lymphadenectomy (Table 2) (25.0% versus 52.8%, P = 0.014). Among the staged cases, the extent of lymphadenectomy was reduced in metachronous cases compared to non-metachronous cases, although this did not reach statistical significance (median number of lymph nodes, 2.5 versus 12, P = 0.057). Radiotherapy was less commonly used for women with metachronous uterine malignancy compared to non-metachronous malignancy (9% versus 29.9%, P = 0.035).
The median follow-up of the censored cases was 6.6 years (interquartile range, 9.3). There were 109,862 survival events recorded during the follow-up period. In the entire cohort, metachronous uterine malignancy was significantly associated with decreased CSS (5-year rates: 48.4% versus 82.2%, HR 3.38, 95%CI 1.76–6.50, P < 0.001; Fig. 3A) and OS (5-year rates: 30.6% versus 73.1%, HR 4.22, 95%CI 2.66–6.70, P < 0.001; Fig. 3B).
Fig. 3.
Survival outcomes of metachronous endometrial cancer. Log-rank test for P-values. (A) cause-specific survival and (B) overall survival are shown between metachronous and non-metachronous endometrial cancer cases.
Tumor factor-specific analysis was performed for survival according to histology, tumor grade, and cancer stage (Table 3). In endometrioid endometrial cancer, OS was significantly worse in metachronous cases compared to non-metachronous cases (HR 6.17, P < 0.001). In non-endometrioid endometrial cancer, OS was marginally worse in metachronous cases compared to non-metachronous cases but the statistical magnitude was smaller than for endometrioid cancers (HR 1.92, P = 0.051).
Table 3.
Survival of metachronous endometrial cancer.
| Cause-specific survival |
Overall survival |
|||
|---|---|---|---|---|
| Characteristic | HR (95%CI) | P-value | HR (95%CI) | P-value |
| Endometrioid | ||||
| Metachronous EMCA (−) | 1 | 1 | ||
| Metachronous EMCA (+) | 4.60 (1.48–14.3) | 0.008 | 6.17 (3.09–12.3) | <0.001 |
| Non-endometrioid | ||||
| Metachronous EMCA (−) | 1 | 1 | ||
| Metachronous EMCA (+) | 1.48 (0.62–3.55) | 0.38 | 1.92 (0.99–3.69) | 0.051 |
| Grade 1–2 | ||||
| Metachronous EMCA (−) | 1 | 1 | ||
| Metachronous EMCA (+) | 2.74 (0.39–19.4) | 0.31 | 7.79 (3.72–16.3) | <0.001 |
| Grade 3 | ||||
| Metachronous EMCA (−) | 1 | 1 | ||
| Metachronous EMCA (+) | 2.00 (0.96–4.20) | 0.07 | 2.15 (1.16–4.00) | 0.015 |
| Stage I–II | ||||
| Metachronous EMCA (−) | 1 | 1 | ||
| Metachronous EMCA (+) | 5.59 (2.10–14.9) | 0.001 | 5.20 (2.71–9.99) | <0.001 |
| Stage III–IV | ||||
| Metachronous EMCA (−) | 1 | 1 | ||
| Metachronous EMCA (+) | 1.57 (0.59–4.19) | 0.37 | 2.29 (1.09–4.81) | 0.028 |
Cox proportional hazard regression model for unadjusted P-values. Significant P-values are emboldened. Abbreviations: HR, hazard ratio; CI, confidence interval; EMCA, endometrial cancer.
Among women with grade 1–2 endometrial cancer, metachronous cases were associated with decreased OS compared to non-metachronous cases (HR 7.79, P < 0.001). In grade 3 endometrial cancers, OS was also worse in metachronous tumors compared to non-metachronous tumors but the statistical impact was smaller than for grade 1–2 tumors (HR 2.15, P = 0.015).
Similarly, in early-stage endometrial cancer, metachronous cases were significantly associated with decreased OS compared to non-metachronous cases (HR for stage I-II: 5.29, P < 0.001). In advanced-stage disease, metachronous cases had decreased OS as well, but the HR was smaller than for early-stage disease (HR for stage III-IV: 2.29, P = 0.028). Similar trends were observed for CSS (Table 3).
4. Discussion
Metachronous uterine malignancy is rare. When these tumors are diagnosed they typically occur many years after the cervical cancer diagnosis. There are few relevant predictors for the development of metachronous uterine malignancy other than low tumor grade. The impact of metachronicity on survival was more important in malignancies with tumor factors that are generally believed to be more favorable, that is low-grade and early-stage disease.
Overall, our study supports the previous literature in that radiation-associated metachronous uterine malignancy is associated with a worse survival compared to non-metachronous tumors [11-13]. However, our study is more specific and informative in that our analysis evaluated tumor factor-specific outcomes related to metachronous uterine malignancy. A previous study attempted this analysis, however the researchers did not proceed due to limited sample size because of rarity of this disease [12]. In our study, adequate numbers in the control group allowed us to perform this analysis, adding new insights in the literature.
Due to lack of prior studies examining the tumor factor-specific significance of metachronous uterine malignancy, it remains unknown as to why metachronicity impacts survival more in tumors that are low-grade, early-stage, and of endometrioid histology compared to tumors that are high-grade, advanced-stage, and of non-endometrioid histology. It is speculated that the following three factors may directly or indirectly influence this association: (i) treatment factors, (ii) host factors, and (iii) tumor factors.
First, our study showed that women with metachronous uterine malignancies were less likely to undergo surgical staging with lymphadenectomy compared to those without metachronous tumors. Similarly, radiotherapy was less likely to be utilized to women with metachronous tumors compared to those without metachronous tumors. Therefore, suboptimal treatment in women with metachronous uterine malignancies could be a causal factor for decreased survival in this disease. This may be particularly applicable in women with early-stage disease.
Women with metachronous uterine malignancy also likely previously received multiple treatments for cervical cancer including pelvic irradiation, and therefore, it is possible that both surgeons and patients may have been reluctant to proceed with aggressive treatments for metachronous uterine malignancy. Radiotherapy-related effects on pelvic tissues may also limit the extent of surgery for metachronous uterine malignancy.
Host factors may also contribute to decreased survival in women with metachronous uterine malignancy. After pelvic irradiation with or without chemotherapy, bone marrow function may be reduced resulting in possible reduced cancer immunity [23,24]. Radiotherapy is generally known, however, to induce enhanced short-term immune function via various mechanisms such as increased immunogenic cell death but the long-term effects are not known [25]. Thus, addressing the long-term effects of radiotherapy on cancer immunity may be of interest in the era of oncoimmunotherapy.
In our study, women who had low-grade cervical tumors had an increased risk of developing metachronous uterine malignancy. This observation is likely due to the fact that women with lower grade cervical cancers have more favorable survival and are more likely to be cured of their disease and become cancer survivors [26], with the potential to develop metachronous uterine malignancy several years remote from their cervical cancer diagnosis.
The effect of pelvic irradiation on the risk of developing uterine sarcoma remains understudied. One review found only sporadic case studies reporting this association [27,28]. There is, however, a large-scale study that reports that the incidence of radiation-induced sarcoma, irrespective of the site of malignancy (gynecologic or non-gynecologic), is <1%, which is reassuring that the risk of sarcoma after radiation is not substantially increased [29]. In our study, the proportion of uterine sarcoma among post-radiation metachronous uterine tumors is similar to non-metachronous uterine sarcoma (4.5% versus 4.3%). Because the sample size is limited in these studies, larger studies are warranted to address this association.
There are a number of limitations of this study. First, this is a retrospective study with possible missing confounders for analysis. For example, information regarding concurrent use of systemic chemotherapy during pelvic radiotherapy is not available in the database. Certain chemotherapy agents are associated with increased risk of secondary primary malignancy, and lack of this information limits our ability to perform risk adjustment. Second, our sample size for metachronous uterine malignancy is low despite a large number of at-risk patients examined in this database. Other studies have reported larger sample sizes of radiation-associated uterine malignancies, however their studies used more heterogeneous study populations, examining malignancies other than cervical cancer [13]. By examining cervical cancer alone with stricter inclusion and exclusion criteria for pelvic irradiation, our results provide more specific information that is applicable in this subset of women with cervical cancer.
Third, the small sample size in our study may partly be due to the relatively short follow-up period (4.8 years for the cervical cancer cohort). The median time to develop radiation-related metachronous uterine malignancy in our study was more than eight years, which is similar to what is reported in the literature. Therefore, a follow-up time of <5 years may not be adequate to capture larger numbers of women with metachronous uterine malignancy, leading to a small event number [11,12].
Fourth, multivariable analysis is generally not feasible in studies of rare tumors. Therefore, while univariable analyses point toward a larger number of metachronous tumors in early-stage/low-grade disease, exact risk adjustment for histology, grade, and stage was not available on multivariable analysis. Lastly, we do not have information regarding a number of important factors that likely influenced treatment and outcome including bone marrow status, performance status, and patient and physician preferences.
Based on our findings, we acknowledge that long-term surveillance to detect metachronous uterine malignancy may not be feasible because of rarity, late onset, and lack of useful clinical and pathological parameter. Nevertheless, both clinicians and patients need to be aware of this rare but highly aggressive secondary primary malignancy, and prompt workup may be warranted when suspected. There are some predictors reported in the literature that can be of use: (i) hematometrea after radiotherapy for cervical cancer may be a sign of metachronous uterine malignancy, and (ii) cervical stenosis is associated with delay in diagnosis [30].
While it is intriguing to consider adjuvant hysterectomy after radiotherapy for locally-advanced cervical cancer to reduce the future risk of metachronous uterine malignancy [31,32], the current guidelines do not support the use of adjuvant hysterectomy in the management of locally-advanced cervical cancer [2,33].
Supplementary Material
HIGHLIGHTS.
Post-radiation metachronous uterine malignancy (MUM) was examined in cervical cancer.
MUM is rare: 10- and 20-year cumulative incidences were 0.6% and 1.2%, respectively.
MUM presents with non-endometrioid, higher grade, and advanced stage disease.
Impact of MUM on survival is more eminent when tumor has favorable factors.
Women with MUM were more likely to have suboptimal treatment.
Acknowledgments
Funding support
Ensign Endowment for Gynecologic Cancer Research (K.M.)
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ygyno.2018.08.035.
Disclosure statement
Consultant for Clovis Oncology and Tesaro (J.D.W.); none for others.
References
- [1].Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A, Global cancer statistics, 2012, CA Cancer J. Clin 65 (2015) 87–108. [DOI] [PubMed] [Google Scholar]
- [2].National Comprehensive Cancer Network Clinical Practice Guideline in Oncology, Cervical cancer, https://www.nccn.org/professionals/physician_gls/pdf/cervical.pdf> (accessed/7/2018).
- [3].Torre LA, Islami F, Siegel RL, Ward EM, Jemal A, Global cancer in women: burden and trends, Cancer Epidemiol. Biomark. Prev 26 (2017) 444–457. [DOI] [PubMed] [Google Scholar]
- [4].Bjorge T, Hennig EM, Skare GB, Soreide O, Thoresen SO, Second primary cancers in patients with carcinoma in situ of the uterine cervix. The Norwegian experience 1970–1992, Int. J. Cancer 62 (1995) 29–33. [DOI] [PubMed] [Google Scholar]
- [5].Arai T, Nakano T, Fukuhisa K, Kasamatsu T, Tsunematsu R, Masubuchi K, Yamauchi K, Hamada T, Fukuda T, Noguchi H, et al. , Second cancer after radiation therapy for cancer of the uterine cervix, Cancer 67 (1991) 398–405. [DOI] [PubMed] [Google Scholar]
- [6].Pettersson F, Ryberg M, Malker B, Second primary cancer after treatment of invasive carcinoma of the uterine cervix, compared with those arising after treatment for in situ carcinomas. An effect of irradiation? A cancer registry study, Acta Obstet. Gynecol. Scand 69 (1990) 161–174. [DOI] [PubMed] [Google Scholar]
- [7].Matsuo K, Machida H, Blake EA, Adams CL, Ciccone MA, Roman LD, Wright JD, Incidence and risk factors of metachronous vulvar, vaginal, and anal cancers after cervical cancer diagnosis, Gynecol. Oncol 150 (3) (2018) 501–508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Rombouts AJM, Hugen N, van Beek JJP, Poortmans PMP, de Wilt JHW, Nagtegaal ID, Does pelvic radiation increase rectal cancer incidence? - a systematic review and meta-analysis, Cancer Treat. Rev 68 (2018) 136–144. [DOI] [PubMed] [Google Scholar]
- [9].Teng CJ, Huon LK, Hu YW, Yeh CM, Chao Y, Yang MH, Chen TJ, Hung YP, Liu CJ, Secondary primary malignancy risk in patients with cervical cancer in Taiwan: a nationwide population-based study, Medicine (Baltimore) 94 (2015), e1803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Boice JD Jr., Day NE, Andersen A, Brinton LA, Brown R, Choi NW, Clarke EA, Coleman MP, Curtis RE, Flannery JT, et al. , Second cancers following radiation treatment for cervical cancer. An international collaboration among cancer registries, J. Natl. Cancer Inst 74 (1985) 955–975. [PubMed] [Google Scholar]
- [11].Pothuri B, Ramondetta L, Martino M, Alektiar K, Eifel PJ, Deavers MT, Venkatraman E, Soslow RA, Barakat RR, Development of endometrial cancer after radiation treatment for cervical carcinoma, Obstet. Gynecol 101 (2003) 941–945. [DOI] [PubMed] [Google Scholar]
- [12].Pothuri B, Ramondetta L, Eifel P, Deavers MT, Wilton A, Alektiar K, Barakat R, Soslow RA, Radiation-associated endometrial cancers are prognostically unfavorable tumors: a clinicopathologic comparison with 527 sporadic endometrial cancers, Gynecol. Oncol 103 (2006) 948–951. [DOI] [PubMed] [Google Scholar]
- [13].Kumar S, Shah JP, Bryant CS, Seward S, Ali-Fehmi R, Morris RT, Malone JM Jr., Radiation-associated endometrial cancer, Obstet. Gynecol 113 (2009) 319–325. [DOI] [PubMed] [Google Scholar]
- [14].The Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute, https://seer.cancer.gov/, Accessed date: 7 July 2018.
- [15].National Cancer Registrars Association, http://www.ncra-usa.org/i4a/pages/index.cfm?pageid=1%3e, Accessed date: 7 July 2018.
- [16].Matsuo K, Machida H, Horowitz MP, Shahzad MMK, Guntupalli SR, Roman LD, Wright JD, Risk of metachronous ovarian cancer after ovarian conservation in young women with stage I cervical cancer, Am. J. Obstet. Gynecol 217 (2017) 580.e1–580.e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Matsuo K, Machida H, Stone RL, Soliman PT, Thaker PH, Roman LD, Wright JD, Risk of subsequent ovarian cancer after ovarian conservation in young women with stage I endometrioid endometrial cancer, Obstet. Gynecol 130 (2017) 403–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Matsuo K, Machida H, Blake EA, Holman LL, Rimel BJ, Roman LD, Wright JD, Trends and outcomes of women with synchronous endometrial and ovarian cancer, Oncotarget 9 (2018) 28757–28771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Kapp DS, Fischer D, Grady KJ, Schwartz PE, Subsequent malignancies associated with carcinoma of the uterine cervix: including an analysis of the effect of patient and treatment parameters on incidence and sites of metachronous malignancies, Int. J. Radiat. Oncol. Biol. Phys 8 (1982) 197–205. [DOI] [PubMed] [Google Scholar]
- [20].Matsuo K, Machida H, Shoupe D, Melamed A, Muderspach LI, Roman LD, Wright JD, Ovarian conservation and overall survival in young women with early-stage low-grade endometrial cancer, Obstet. Gynecol 128 (2016) 761–770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Matsuo K, Machida H, Shoupe D, Melamed A, Muderspach LI, Roman LD, Wright JD, Ovarian conservation and overall survival in young women with early-stage cervical cancer, Obstet. Gynecol 129 (2017) 139–151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Matsuo K, Machida H, Frimer M, Marcus JZ, Pejovic T, Roman LD, Wright JD, Prognosis of women with stage I endometrioid endometrial cancer and synchronous stage I endometrioid ovarian cancer, Gynecol. Oncol 147 (2017) 558–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Newman NB, Sidhu MK, Baby R, Moss RA, Nissenblatt MJ, Chen T, Lu SE, Jabbour SK, Long-term bone marrow suppression during postoperative chemotherapy in rectal cancer patients after preoperative chemoradiation therapy, Int. J. Radiat. Oncol. Biol. Phys 94 (2016) 1052–1060. [DOI] [PubMed] [Google Scholar]
- [24].Mauch P, Constine L, Greenberger J, Knospe W, Sullivan J, Liesveld JL, Deeg HJ, Hematopoietic stem cell compartment: acute and late effects of radiation therapy and chemotherapy, Int. J. Radiat. Oncol. Biol. Phys 31 (1995) 1319–1339. [DOI] [PubMed] [Google Scholar]
- [25].Walle T, Martinez Monge R, Cerwenka A, Ajona D, Melero I, Lecanda F, Radiation effects on antitumor immune responses: current perspectives and challenges, Ther. Adv. Med. Oncol 10 (2018), 1758834017742575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Matsuo K, Mandelbaum RS, Machida H, Purushotham S, Grubbs BH, Roman LD, Wright JD, Association of tumor differentiation grade and survival in women with squamous cell carcinoma of the uterine cervix, J. Gynecol. Oncol (2018) accepted. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Marmion PJ, Goldfarb PM, Youngkin TP, Uterine sarcoma following adjuvant radiotherapy for rectal carcinoma, J. Surg. Oncol 17 (1981) 63–67. [DOI] [PubMed] [Google Scholar]
- [28].Morrel B, Mulder AF, Chadha S, Tjokrowardojo AJ, Wijnen JA, Angiosarcoma of the uterus following radiotherapy for squamous cell carcinoma of the cervix, Eur. J. Obstet. Gynecol. Reprod. Biol 49 (1993) 193–197. [DOI] [PubMed] [Google Scholar]
- [29].Mark RJ, Poen J, Tran LM, Fu YS, Selch MT, Parker RG, Postirradiation sarcomas. A single-institution study and review of the literature, Cancer 73 (1994) 2653–2662. [DOI] [PubMed] [Google Scholar]
- [30].Lopez J, Horta M, Cunha TM, Endometrial cancer after radiation therapy for cervical carcinoma: a radiological approach, Eur. J. Radiol 105 (2018) 283–288. [DOI] [PubMed] [Google Scholar]
- [31].Keys HM, Bundy BN, Stehman FB, Muderspach LI, Chafe WE, Suggs CL 3rd, Walker JL, Gersell D, Cisplatin, radiation, and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage IB cervical carcinoma, N. Engl. J. Med 340 (1999) 1154–1161. [DOI] [PubMed] [Google Scholar]
- [32].Keys HM, Bundy BN, Stehman FB, Okagaki T, Gallup DG, Burnett AF, Rotman MZ, Fowler WC Jr., Radiation therapy with and without extrafascial hysterectomy for bulky stage IB cervical carcinoma: a randomized trial of the Gynecologic Oncology Group, Gynecol. Oncol 89 (2003) 343–353. [DOI] [PubMed] [Google Scholar]
- [33].Shim SH, Kim SN, Chae SH, Kim JE, Lee SJ, Impact of adjuvant hysterectomy on prognosis in patients with locally advanced cervical cancer treated with concurrent chemoradiotherapy: a meta-analysis, J. Gynecol. Oncol 29 (2018), e25. [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.