Effects of radiotherapy on the survival of patients with stage IA and low-grade stage IB uterine endometrioid carcinoma

The present study aimed to evaluate the effects of radiotherapy on the overall survival of patients with primary stage IA, grade I–III uterine endometrioid carcinoma or stage IB, grade I–II uterine endometrioid carcinoma. A total of 7504 patients with stage IA, grade I–III uterine endometrioid carcinoma, and 857 patients with stage IB, grade I–II uterine endometrioid carcinoma were collected for the present study. Following propensity score matching (PSM), statistical analysis was performed for the equalized number of patients with stage IA, grade I–III uterine endometrioid carcinoma (n = 383) and patients with stage IB, grade I–II uterine endometrioid carcinoma (n = 330). For patients with primary stage IA, grade I–III uterine endometrioid carcinoma, radiotherapy was found to promoted a reduced 5-year overall survival rates [hazard ratio (HR), 1.726; 95% confidence interval (CI), 1.456–2.046; P < 0.05]. In patients with primary stage IB, grade I–II uterine endometrioid carcinoma, no significant differences were observed in the 5-year overall survival rates between radiotherapy and no radiotherapy groups (P = 0.059). In conclusion, radiotherapy may not improve 5-year overall survival for patients with primary stage IA, grade I–III or stage IB, grade I–II uterine endometrioid carcinoma.


Inclusion criteria
Patients with primary stage IA, grade I-III uterine endometrioid carcinoma or stage IB, grade I-II uterine endometrioid carcinoma were identified.Patients were included in the present study following surgery, when chemotherapy had not been previously conducted.Information regarding tumor size and the number of lymph nodes removed was required.Notably, patients included in the present study were not positive for pelvic lymph nodes or para-aortic lymph nodes.Diagnosis depended on cytology or histology.
A total of 7504 patients with stage IA, grade I-III uterine endometrioid carcinoma and 857 patients with stage IB, grade I-II uterine endometrioid carcinoma were selected for inclusion in the present study.The following variables were measured: Radiotherapy, age, race, tumor grade, tumor size, regional lymph nodes removed and survival.In the extracted dataset, radiotherapy was categorized into two groups: (i) Yes and (ii) no; age was divided into two groups: (i) < 60 and (ii) ≥ 60 years; race was classified into two groups: (i) white and (ii) nonwhite; grade was categorized into three groups: (i) well differentiated, grade I; (ii) moderately differentiated, grade II; and (iii) poorly differentiated, grade III.In addition, tumor size was categorized into two groups: (i) < 2 and (ii) ≥ 2 cm; and regional lymph nodes were classified into two groups: (i) none and (ii) removed.All subjects were assessed for the association between radiotherapy and 5-year overall survival (Figs. 1 and 2).

Statistical analysis
SPSS software (version 26; IBM Corp.) was used to perform statistical analysis.Univariate associations between categorical variables and radiotherapy were assessed using Pearson's χ 2 -tests.All statistical tests were two-sided.P < 0.05 was considered to indicate a statistically significant difference.Propensity score matching (PSM) was used to minimize selection bias, which is common and inevitable in retrospective studies.Categorical variables included age, race, grade, tumor size and regional lymph nodes removed in both radiotherapy and no radiotherapy groups, and these were matched using a 1:1 match ratio method.Survival analysis was performed using the Kaplan-Meier method and log-rank tests.The Cox proportional hazards models were used to estimate the hazard ratio (HR) and 95% confidence interval (CI).

Demographic features
There were 7504 patients (7119 who had not received radiotherapy and 385 who had) with primary stage IA, grade I-III uterine endometrioid carcinoma who met the study inclusion criteria.Among these patients, 49.5% were aged < 60 years and 50.5% were aged ≥ 60 years.A total of 84.2% of these patients were white.In addition, 74.4% of the patients were grade I, 22.9% were grade II and only 2.7% were grade III.A total of 38.8% of the patients had tumors measuring < 2 cm in size, whereas 61.2% had tumors measuring ≥ 2 cm.Regional lymph nodes were removed from only 4.7% of the patients, and 5-year overall survival rates were 72.6% (Table 1).
There were 857 patients (497 who had not received radiotherapy and 360 who had) with primary stage IB, grade I-II uterine endometrioid carcinoma who met the study inclusion criteria.The majority of patients were aged ≥ 60 years (n = 660; 77.0%), and 87.7% of the patients were white.A total of 55.8% of the patients were grade I and 44.2% were grade II.Only 10.6% of the patients had tumors measuring < 2 cm in size, whereas 89.4% had tumors measuring ≥ 2 cm.Regional lymph nodes were removed from only 11.6% of the patients, and 5-year overall survival rates were 63.9% (Table 2).

Comparison of univariate categorical variables
The majority of patients with primary stage IA, grade I-III uterine endometrioid carcinoma in the radiotherapy groups were older than those in the no radiotherapy groups (< 60 years, 38.2 vs. 50.1%,respectively; and ≥ 60 years, 61.8 vs. 49.9%,respectively; P < 0.05).A higher number of patients were white in the radiotherapy groups (P = 0.405).A higher number of patients were grade II and grade III in the radiotherapy groups (grade I, 33.5 vs. 76.6%,respectively; grade II, 48.8 vs. 21.5%,respectively; and grade III, 17.7 vs. 1.8%, respectively; P < 0.05).Fewer patients had tumors measuring < 2 cm and an increased number of patients had tumors measuring ≥ 2 cm in the radiotherapy groups (< 2 cm, 15.8 vs. 40.0%,respectively; and ≥ 2 cm, 84.2 vs. 60.0%,respectively; P < 0.05).In addition, there were more patients with regional lymph nodes removed in the radiotherapy groups compared with the no radiotherapy group (11.2 vs. 4.3%, respectively; P < 0.05).Notably, selection bias and non-uniformity between the radiotherapy and no radiotherapy groups were eliminated using PSM.Statistical analysis was further performed for the equalized number of patients (n = 383).The categorical variables were well-balanced and the potential covariates between the two groups were markedly decreased (Table 3).
Fewer elderly patients with primary stage IB, grade I-II uterine endometrioid carcinoma tended to receive radiotherapy (< 60 years, 25.8 vs. 20.9%,respectively; and ≥ 60 years, 74.2 vs. 79.1%,respectively; P < 0.05).A higher number of patients were white in the radiotherapy groups (P = 0.851).Fewer patients were grade I, and an increased number of patients were grade II in the radiotherapy groups, compared with those in the no radiotherapy groups (grade I, 46.7 vs. 62.4%, respectively; and grade II, 53.3 vs. 37.6%, respectively; P < 0.05).Fewer patients had tumors measuring < 2 cm in size and an increased number of patients had tumors measuring ≥ 2 cm in the radiotherapy groups (< 2 cm, 6.1 vs. 13.9%,respectively; and ≥ 2 cm, 93.9 vs. 86.1%,respectively; P < 0.05).Fewer patients had regional lymph nodes removed in the radiotherapy groups compared with the no radiotherapy groups (P = 0.899).All categorical variables were well-balanced without significant differences following PSM.An equal number of patients were further analyzed (n = 330; Table 4).

Univariate analysis of categorical variables and overall survival
The association between univariate categorical variables and 5-year overall survival rates was analyzed using the Kaplan-Meier method and log-rank tests.For patients with primary stage IA, grade I-III uterine endometrioid carcinoma, radiotherapy decreased 5-year overall survival rates compared with the no radiotherapy groups (69.5 vs. 81.7%,respectively; P < 0.05).There were no significant differences in 5-year overall survival rates between the < 60 and ≥ 60 years age groups (78.1 vs. 74.0%,respectively; P = 0.122).Grade was a risk factor for 5-year overall survival rates (grade I, 82.0%; grade II, 75.0%; and grade III, 64.1%, respectively; P < 0.05).Tumors < 2 cm in size were associated with higher 5-year overall survival rates compared with the tumor ≥ 2 cm groups (81.5  www.nature.com/scientificreports/ vs. 74.2%,respectively; P < 0.05).Notably, race and the removal of regional lymph nodes did not affect 5-year overall survival rates (P = 0.593 and P = 0.479; respectively; Table 5).
For patients with primary stage IB, grade I-II uterine endometrioid carcinoma, 5-year overall survival rates were comparable between radiotherapy and no radiotherapy groups (64.5 vs. 65.2%,respectively; P < 0.05).Notably, age was a risk factor for 5-year overall survival rates, and these were 69.8 and 63.4% in < 60 and ≥ 60 years age groups, respectively (P < 0.05).There were no significant differences in 5-year overall survival rates between the grade I and grade II groups (71.1 vs. 58.3%,respectively; P = 0.446).In addition, tumors < 2 cm in size were associated with a higher 5-year overall survival rate, compared with the tumor ≥ 2 cm groups (77.8 vs. 63.9%,respectively; P < 0.05).Race and the removal of regional lymph nodes did not affect 5-year overall survival rates (P = 0.346 and P = 0.562; respectively; Table 6).
To investigate how the categorical variables impacted 5-year overall survival in combination, these factors were incorporated into the multivariate Cox regression analysis.For patients with primary stage IA, grade I-III uterine endometrioid carcinoma, radiotherapy decreased 5-year overall survival rates (HR, 1.726; 95% CI,  9).For patients with primary stage IB, grade I-II uterine endometrioid carcinoma, there were no significant differences in 5-year overall survival rates between the radiotherapy and no radiotherapy groups (P = 0.059).Increased age was associated with reduced 5-year overall survival rates (HR, 1.300; 95% CI, 1.040-1.623;P < 0.05).In addition, 5-year overall survival rates were reduced in the tumor ≥ 2 cm groups (HR, 1.455; 95% CI, 1.025-2.063;P < 0.05; Table 10).

Discussion
Results of the present study demonstrated that 5-year overall survival was not improved following radiotherapy in patients with primary stage IA, grade I-III or stage IB, grade I-II uterine endometrioid carcinoma, who underwent surgery with no subsequent chemotherapy, and exhibited no positive pelvic lymph nodes or paraaortic lymph nodes.In accordance with results of the present study, Creutzberg et al. demonstrated that postoperative radiotherapy exerted no significant impact on overall survival in patients with stage IA and stage IB endometrial carcinoma 15 .Similarly, the results of previous randomized trials demonstrated that radiotherapy reduced the risk of local-regional recurrence, but did not improve overall survival for patients with early-stage endometrial carcinoma [16][17][18] .
The management of radiotherapy in early-stage endometrial carcinoma remains controversial.Rydzewski et al. reported that radiotherapy was associated with a reduction in overall mortality and an improvement in survival for patients with stage IA and IB endometrial carcinoma 19 .Similarly, results of a previous study demonstrated that radiotherapy significantly increased the survival of patients with stage IA and IB endometrial carcinoma 20 .Medenwald et al. revealed that radiotherapy in addition to surgery is beneficial for patients with stage IB endometrial carcinoma, but not for stage IA endometrial carcinoma 21 .These discrepancies may be due to study design, prediction of disease recurrence or prognosis.Moreover, these previous studies not only investigated endometrioid histology, but also other histological types.Although the majority of endometrial carcinomas include endometrioid histology, conclusions drawn in the present study cannot be applied to all patients with endometrioid histology.A stratified subgroup analysis of histological subtypes is required, and may demonstrate distinct differences in survival.In addition, previous studies including patients who received radiotherapy may lead to selection bias.Patients receiving radiotherapy may have improved health and financial status, which may impact survival time.Moreover, previous data was discounted in previous studies due to confounding factors,  www.nature.com/scientificreports/and the sample sizes analyzed were small; thus, statistical analyses may be impacted.Studies with improved designs and large-scale clinical trials are required to address the aforementioned limitations.In certain studies, the risk of disease recurrence could not be reliably predicted in early-stage endometrial carcinomas due to clinicopathologic features, such as histology, stage, grade or lymph node involvement; thus, these results may not be reproducible 22,23 .Therefore, the management of early-stage endometrial carcinomas remains controversial.
Results of the present study demonstrated that in patients with primary stage IA, grade I-III uterine endometrioid carcinoma, 5-year overall survival was markedly reduced by grade II/III and tumors ≥ 2 cm in size.Moreover, 5-year overall survival was significantly decreased following radiotherapy, not prospectively increased.For patients with primary stage IB, grade I-II uterine endometrioid carcinoma, age ≥ 60 years and tumors ≥ 2 cm in size were risk factors, which was consistent with the results of previous studies, and these factors reduced 5-year overall survival.Notably, no significant differences in 5-year overall survival were observed between the radiotherapy and no radiotherapy groups.Therefore, radiotherapy was not an appropriate treatment to improve 5-year overall survival for patients with primary stage IA, grade I-III or stage IB, grade I-II uterine endometrioid carcinoma, who underwent surgery and did not receive chemotherapy, and exhibited no positive pelvic lymph nodes or para-aortic lymph nodes.The aforementioned conclusion was drawn following the exclusion of all confounding factors.Notably, the risk of recurrence for early-stage uterine endometrioid carcinoma was low and prognosis was favorable.In addition, the side effects of radiotherapy may affect the survival of patients.Therefore, results of the present study may be encouraging.
The present study was based on a large-scale and comprehensive population-based cohort, and provided evidence on the efficacy of radiotherapy for treating primary stage IA, grade I-III and stage IB, grade I-II uterine endometrioid carcinoma.In addition, all subjects were randomly enrolled without subjective preference, which resulted in an objective conclusion.
However, the present study exhibits limitations.Notably, the SEER database lacked detailed information regarding lymph-vascular invasion.However, the risk of lymph-vascular invasion is low for early-stage endometrioid carcinoma 24 .Moreover, PSM was used to randomize the data and to decrease selection bias caused by this limitation.
In conclusion, radiotherapy may not improve 5-year overall survival for patients with primary stage IA, grade I-III or stage IB, grade I-II uterine endometrioid carcinoma.Further clinical trials are required to verify the conclusions of the present study. https://doi.org/10.1038/s41598-023-46435-ywww.nature.com/scientificreports/

Figure 1 .
Figure 1.Flow chart for patients with primary stage IA, grade I-III uterine endometrioid carcinoma.

Figure 2 .
Figure 2. Flow chart for patients with primary stage IB, grade I-II uterine endometrioid carcinoma.

Table 1 .
Demographics for patients with stage IA uterine endometrioid carcinoma.

Table 3 .
Comparison of univariate covariates for patients with stage IA uterine endometrioid carcinoma.

Table 4 .
Comparison of univariate covariates for patients with stage IB uterine endometrioid carcinoma.PSM propensity score matching.P-value < 0.05 is regarded as statistically significant.

Table 5 .
Univariate analysis of clinical factors with 5-year overall survival for patients with stage IA uterine endometrioid carcinoma.

Table 6 .
Univariate analysis of clinical factors with 5-year overall survival for patients with stage IB uterine endometrioid carcinoma.

Table 7 .
Univariate cox regression analysis for 5-year overall survival for patients with stage IA uterine endometrioid carcinoma.

Table 9 .
Multivariate cox regression analysis for 5-year overall survival for patients with stage IA uterine endometrioid carcinoma.