Clinical outcomes analysis of image-guided brachytherapy as definitive treatment for inoperable endometrial cancer

Abstract

Objectives

This study evaluates the efficacy and toxicity of image-guided brachytherapy combined with or without external beam radiotherapy (IGBT ± EBRT) as definitive treatment for patients with inoperable endometrial cancer (IOEC), in addition to establishing a risk classification to predict prognosis.

Methods

Fifty-one IOEC patients who underwent IGBT ± EBRT at Peking Union Medical College Hospital from January 2012 to December 2021 were retrospectively analyzed, of which 42 patients (82.4%) were treated with IGBT + EBRT and 9 patients (17.6%) with IGBT alone. Establishing risk classification based on FIGO 2009 staging and biopsy pathology, stage III/IV, non-endometrioid, or Grade 3 endometrioid cancer were included in the high-risk group (n = 25), and stage I/II with Grade 1–2 endometrioid cancer was included in the low-risk group (n = 26).

Results

The median follow-up time was 58.0 months (IQR, 37.0–69.0). Clinical complete remission (CR) was achieved in 92.2% of patients after radiotherapy (n = 47). The cumulative incidences of locoregional and distant failure were 19.6% (n = 10) and 7.8% (n = 4), respectively. A total of 20 patients died (39.2%), including 10 cancer-related deaths (19.6%) and 10 comorbidity-related deaths (19.6%). The 5-year locoregional control (LRC), time to progression (TTP), overall survival (OS), and cancer-specific survival (CSS) were 76.9%, 71.2%, 59.4%, and 77.0%, respectively. No Grade 3 or above acute or late toxicities were reported. In univariate analysis, LRC, TTP, and CSS were significantly higher in the low-risk group than in the high-risk group (P < 0.05). After adjusting for age, number of comorbidities, radiotherapy modality, and chemotherapy, the low-risk group was still significantly better than the high-risk group in terms of LRC (HR = 6.10, 95% CI: 1.18–31.45, P = 0.031), TTP (HR = 8.07, 95% CI: 1.64–39.68, P = 0.010) and CSS (HR = 6.29, 95% CI: 1.19–33.10, P = 0.030).

Conclusions

IGBT ± EBRT is safe and effective as definitive treatment for IOEC patients, achieving satisfactory locoregional control, favorable survival outcomes, and low toxicity. Risk classification based on FIGO 2009 staging and biopsy pathology is an independent prognostic factor for LRC, TTP, and CSS.

Introduction

Endometrial carcinoma (EC) is the most prevalent malignant tumor of the female reproductive system in developed countries. In China, it has become the second most common gynecological cancer, with an incidence of 10.54/100,000 in 2016, especially higher in urban areas [1]. The standard treatment for EC is total hysterectomy and bilateral salpingo-oophorectomy with or without lymph node dissection. Adjuvant therapy, including pelvic radiotherapy, brachytherapy, and systemic therapy, can be selected based on individual clinical and pathological high-risk factors [2, 3]. Currently, a small percentage of EC patients (3–10%) may be unable to tolerate surgery due to multiple medical comorbidities [4]. Advanced age, obesity, hypertension, and diabetes are all closely associated with an increased risk of EC. With the aging of the population and the increasing prevalence of obesity and chronic diseases, the number of patients with inoperable endometrial cancer (IOEC) is growing [5, 6]. In addition, some patients with advanced-stage disease at diagnosis may also be inoperable due to the difficulty of local tumor resection. For these patients who are not candidates for surgery, definitive radiotherapy, namely brachytherapy combined with or without external beam radiotherapy (EBRT), can be used as an effective treatment option, particularly for improving local control. A systematic review has shown that the 5-year cancer-specific survival and 5-year local control of inoperable patients who receive definitive radiotherapy were 78.5% and 79.9%, respectively [7]. However, the majority of historical studies were based on outdated radiation techniques and focused on early-stage patients. Large-scale retrospective cohort studies and prospective cohort studies are still lacking to support the clinical efficacy and safety of definitive radiotherapy for IOEC patients.

With the rapid development of radiotherapy technology in the past decades, local control and survival have been significantly improved, while the toxicity has been considerably reduced. In terms of EBRT, intensity-modulated radiotherapy (IMRT) significantly reduced late small bowel and bladder complications compared with conventional three-dimensional conformal radiotherapy (3D-CRT) [8,9,10]. In terms of brachytherapy, the traditional low-dose rate (LDR) 2D brachytherapy has been converted to high-dose-rate (HDR) 3D image-guided brachytherapy (IGBT). The use of CT or MRI-based 3D image-guiding has improved the accuracy of clinical target volume (CTV) delineation, thereby ensuring sufficient dose coverage to the CTV while minimizing radiation damage to surrounding normal organs [11, 12]. Therefore, summarizing the clinical outcomes of IOEC patients treated with modern radiation techniques is crucial to update our understanding of the efficacy of radiotherapy as a definitive treatment for IOEC patients.

Currently, risk classification is routinely applied in the postoperative adjuvant treatment selection and prognostic assessment of endometrial cancer patients. There are various clinical trials or guidelines put forward different risk classification criteria up to now, including the GOG (Gynecologic Oncology Group), PORTEC (Postoperative Radiotherapy for Endometrial Cancer), and ESMO-ESGO-ESTRO (European Society for Medical Oncology, European Society of Gynaecological Oncology, European Society for Radiotherapy and Oncology) [2, 13]. All the risk classifications mentioned above were based on postoperative pathological results, and the risk factors included age, stage, grade, depth of myometrial invasion, the status of lymph-vascular space invasion (LVSI), etc. However, conventional risk classification is not applicable to IOEC patients due to the lack of surgical pathological results, which makes it difficult to achieve precise treatment management and prognosis evaluation. Therefore, it is essential to establish an appropriate risk classification for this specific group of patients.

Although several previous studies have investigated HDR-IGBT for IOEC patients, most of them are based on European and American populations with relatively small sample sizes. Moreover, there are few analyses of the factors affecting prognosis, and no effective risk classification has been established to predict the prognosis. The current study is a single-center retrospective cohort study with the largest Asian cohort currently available, aiming to verify the efficacy and toxicity of IGBT ± EBRT for IOEC patients and establish a reasonable risk classification to help them evaluate the treatment prognosis.

Materials and methods

Patient population

The medical records of patients with IOEC who received IGBT ± EBRT as definitive treatment in Peking Union Medical College Hospital from January 2012 to December 2021 were retrospectively analyzed. IOEC was defined as endometrial cancer that is inoperable either due to medical reasons (anesthesia contraindications because of comorbidities) or surgical reasons (extensive invasion preventing tumor resection), which is determined by the evaluation of gynecologists and anesthesiologists. Patients with less than 3 months follow-up time, incomplete survival information, and those treated with 2D brachytherapy techniques or EBRT alone were excluded. Each patient had a histologically confirmed diagnosis of endometrial cancer by biopsy. Staging examinations included pelvic physical examination, transvaginal ultrasound, CT, MRI, and PET-CT. The following information was collected from medical records: (1) personal information, including age, body mass index (BMI), and comorbidity; (2) clinical and pathological characteristics, including stage, histological type, grade, method of staging examination, radiotherapy mode and dosimetric parameters, and chemotherapy; (3) treatment outcome, including toxicity, recurrence, metastasis, and death.

Treatment

For FIGO 2009 Stage IA EC patients with histological Grades 1–2 and small tumor volume, IGBT was delivered alone. On the other hand, patients with higher stages, grades, or larger tumor volumes underwent a combination of EBRT and IGBT.

EBRT was delivered with 6 MV linear accelerators using IMRT techniques, including fixed-field IMRT, volumetric modulated arc therapy (VMAT), and tomotherapy. EBRT fields included the pelvis +/- para-aortic lymph nodes (in cases of para-aortic lymph nodes involvement according to MRI, PET-CT or CT-scan). The prescription dose for EBRT was 45–50.4 Gy in 25–28 fractions (1.8 Gy per fraction), with a simultaneous boost of 60.2 Gy for confirmed metastatic lymph nodes. The CTV encompassed the entire uterus, cervix, upper 1/2 of the vagina, parametrial tissue, and regional lymph drainage areas (obturator, presacral, internal iliac, external iliac, and common iliac nodes). Treatment was delivered with a comfortably filled bladder and an empty rectum. Considering the potiential inter-fractional variation in bladder and rectal filling, the anterior 1/3 of the mesorectum and the posterior 1/4 of the bladder were also included in the CTV. If para-aortic lymph nodes were treated, CTV was extended 1–1.5 cm above the renal artery and at least 2 cm above the highest pathological lymph node. The planning target volume (PTV) was obtained by 6–10 mm margin expansion based on the CTV. The plan optimization parameters require 95% (D95) of the PTV to receive 100% of the prescription dose.

Intracavitary brachytherapy was performed with an iridium-192 source using the HDR afterloading system (Elekta, Sweden), following EBRT or as monotherapy. A variety of applicators were chosen to accommodate each patient’s anatomy, including single tandem with ovoids and single tandem with vaginal cylinder (Nucletron, Elekta, Sweden). Once the applicator was inserted, vaginal packing was used to displace vaginal mucosa, bladder, and rectum, and to hold the applicator in place. CT imaging was performed after each insertion for the 3D treatment planning. The high-risk CTV (HR-CTV) for brachytherapy encompasses the whole uterus and cervix for Stage IA Grade 1–2 patients, whereas it encompasses the whole uterus, cervix and the upper 1–2 cm of the vagina for other patients. A single-fraction prescription dose of 500–700 cGy was delivered to 90% (D90) of the HR-CTV over 4–6 fractions, with 2 fractions per week.

For comparison purposes we calculate the equivalent dose in 2 Gy fractions (EQD2), with the assumed α/β ratio of 10 Gy for endometrial cancer and 3 Gy for organs at risk (OARs). The planning goal for patients treated with IGBT alone was to deliver a D90 EQD2 of 40–48 Gy to the HR-CTV; while for those receiving both EBRT and IGBT, the goal was to achieve a total D90 EQD2 of 70–75 Gy to the HR-CTV. Delineated OARs included the rectum, bladder, and sigmoid colon and small intestine. The minimum dose received by maximally irradiated 2 cm³ volume (D2cc) for OAR were constrained; optimization goals were as follows: rectum D2cc EQD2 ≤ 75 Gy, bladder D2cc EQD2 ≤ 85–90 Gy, sigmoid colon D2cc EQD2 ≤ 75 Gy, small intestine D2cc EQD2 ≤ 65 Gy. Priority was placed on meeting critical organ constraints and thus, when needed, CTV coverage was compromised. Treatment plans were generated using Oncentra Brachy Planning System (Elekta, Sweden) and were manually optimized with each application to obtain the best possible dose coverage of target volume and avoid critical structures. An example treatment plan demonstrating target and organ-at-risk contouring in axial and sagittal views is included in Fig. 1.

In terms of systemic therapy, combined chemotherapy is not indicated for patients with disease confined to the uterus. For those with cervical involvement, concurrent chemoradiotherapy with weekly cisplatin as a sensitizer was considered. In patients with advanced disease (those with extrauterine disease), concurrent chemotherapy followed by adjuvant chemotherapy was recommended, with paclitaxel and carboplatin as the preferred regimen of adjuvant chemotherapy. Chemotherapy could be prescribed based on a multidisciplinary team’s assessment, taking into account the patient’s comorbidities, tumor stage, histological type, and personal preferences to tailor the most appropriate treatment.

Acute and late toxicity related to treatment were retrospectively graded using the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. Adverse events occurring later than 90 days from completion of radiotherapy were defined as late toxicities.

Fig. 1

A 3D-image guided brachytherapy plan for an IOEC patient with FIGO 2009 stage IA and Grade 2. The contouring data is shown in axial (A) and sagittal (C) slices of computed tomography-scan. Red: HR- CTV. It included the whole uterine body and the cervix. Orange: rectum. Yellow: bladder. Blue: sigmoid colon. Green: small intestine. Isodose distributions are depicted in axial (B) and sagittal (D) slices of computed tomography-scan. The 150%, 100%, and 80% isodose lines outlined in green, yellow, and pink, respectively

Follow up

To evaluate the treatment response of patients, pelvic MRI, CT, ultrasound, and/or PET-CT examination was performed 1–3 months after radiotherapy. Disappearance of all tumor lesions, shrinkage of metastatic lymph nodes to normal size, and absence of new lesions were defined as clinical complete remission (CR), which was preferably assessed by pelvic MRI, and in patients with contraindications to MRI by ultrasound, CT and/or PET-CT. Thereafter, patients were assessed every 3–6 months for the first 2 years after radiotherapy, every 6–12 months during the following three years, and then annually. Each evaluation included pelvic physical examination, pelvis MRI, CT, and/or ultrasound. If the possibility of local recurrence was suspected by imaging, it was further confirmed by biopsy. Chest and abdominal CT were performed annually to rule out distant metastasis.

Definition

Risk classification

According to the FIGO 2009 staging system, patients were classified as early-stage disease (stage I or II) or advanced-stage disease (stage III or IV). Pathological types that are non-endometrioid carcinoma (including serous carcinoma, neuroendocrine carcinoma, carcinosarcoma, and mixed carcinoma in our cohort) or Grade 3 endometrioid carcinoma were defined as “invasive pathology types”. High-risk group was defined as either advanced-stage (stage III or IV) or invasive pathology type; low-risk group was defined as early-stage disease (stage I or II) with Grade 1–2 endometrioid carcinoma.

Survival outcomes

All patient survivals were calculated from the end of radiotherapy. The endpoint event of overall survival (OS) was defined as death from any cause or the last follow-up. The endpoint event of cancer-specific survival (CSS) was defined as cancer-related death or the last follow-up. The endpoint event of time to progression (TTP) was defined as any pattern of treatment failure (including locoregional and distant failure) or the last follow-up. The endpoint event of locoregional control (LRC) was defined as locoregional failure or the last follow-up. Locoregional failure was defined as the absence of CR after treatment, local recurrence, and pelvis lymph node metastasis.

Data analysis

Medians and interquartile ranges (IQR) were described for continuous variables, and the significance of differences between the two groups was assessed using the Wilcoxon rank test. Numbers and percentages were described for categorical variables. The statistical significance in a 2 × 2 table was evaluated using Pearson’s chi-squared test or Fisher’s exact test. The survival analysis estimated OS, CSS, TTP, and LRC using the Kaplan–Meier method. The log-rank test was used for the univariate analysis to analyze the survival difference between groups. The Cox proportional hazards regression model was used for the multivariate analysis to calculate the adjusted hazard ratio (HR) and 95% confidence interval (95% CI). A two-tailed test was used, and a P value of < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 26 (IBM Corp., Armonk, New York), R version 4.3.0 (R Core Team. R Foundation for Statistical Computing, Vienna, Austria), and GraphPad Prism 9 (GraphPad Software, La Jolla, California, USA).

Results

Patient and tumor characteristics

From January 2012 to December 2021, 51 patients with IOEC who received IGBT ± EBRT in Peking Union Medical College Hospital were enrolled in this study. Table 1. summarizes the clinical and treatment characteristics of all patients. The median age of the whole cohort was 68.0 years (IQR, 55.0–79.0), and the median BMI was 26.8 kg/m² (IQR, 22.0-29.2). Out of the 51 patients, 45 cases (88.2%) were ineligible for surgery due to the presence of multiple medical comorbidities. And in six cases (11.8%), extensive tumor lesions made surgical removal challenging. Hypertension, heart disease, and diabetes were the most common comorbidities in IOEC patients, which were reported in 34 patients (66.7%), 25 patients (49.0%), and 20 patients (39.2%), respectively. In addition, the prevalence of cerebrovascular disease, chronic kidney disease, pulmonary disease, and liver cirrhosis in IOCE patients ranged from 5 to 20% (Table S1). More than one-third of the patients (20 cases, 39.2%) had three or more comorbidities.

According to FIGO 2009 staging system, 31 patients (60.8%) were classified as stage I, 4 patients (7.8%) as stage II, 12 patients (23.5%) as stage III, and 6 patients (11.8%) as stage IV. The pathological type of most patients was endometrioid carcinoma, accounting for 45 cases (88.2%). There were only 6 cases (11.8%) of non-endometrioid endometrial cancer, including 1 serous carcinoma, 1 carcinosarcoma, 1 neuroendocrine carcinoma, and 3 mixed carcinomas. Grades 1, 2, 3, and unknown were diagnosed in 25 patients (49.0%), 12 patients (23.5%), 12 patients (23.5%), and 2 patients (3.9%), respectively. Regarding staging examination, 41 patients (80.4%) underwent MRI, while the remaining 10 patients (19.6%) received CT and/or PET-CT due to MRI contraindications.

Based on the risk classification defined in this study, 26 patients (51.0%) were classified as low-risk patients and 25 patients (49.0%) as high-risk patients. There is a statistically significant difference in the treatment modalities between the two risk groups (χ² = 10.508, P = 0.002 for radiotherapy mode and χ² = 8.359, P = 0.005 for chemotherapy). High-risk group tended to receive more aggressive treatment approaches than the low-risk group, with all of them undergoing IGBT + EBRT and 9 cases (36.0%) receiving chemotherapy. In addition, patients in the high-risk group had better underlying conditions, with fewer comorbidities compared to the low-risk group (χ² = 4.763, P = 0.045).

Table 1 Clinical and treatment characteristics for low-risk and high-risk patients

Treatment and toxicity

A summary of the radiotherapy dosimetric parameters grouped by radiotherapy modalities is presented in Table 2. 9 patients (17.6%) received IGBT alone with a single fraction dose of 600 cGy delivered over 5–6 fractions. For target coverage, the median HR-CTV D90 EQD2 was 36.2 Gy (IQR 34.2–39.9 Gy), and HR-CTV V100 was 80.0% (IQR 74.5–86.5%). The median D2cc EQD2 for OARs were 15.5 Gy (IQR 11.0–17.9 Gy) for the rectum, 16.3 Gy (IQR 13.2–22.1 Gy) for the bladder, 15.9 Gy (IQR 13.3–21.3 Gy) for the sigmoid colon, and 20.1 Gy (IQR 18.1–24.6 Gy) for small intestine.

42 patients (82.4%) received EBRT with 45.0–50.4 Gy in 25–28 fractions, followed by IGBT with a single fraction dose of 500–700 cGy over 4–6 fractions. For target coverage, the median HR-CTV D90 EQD2 was 79.9 Gy (IQR 73.0–81.8 Gy), and HR-CTV V100 was 79.0% (IQR 68.8–83.3%). The median D2cc EQD2 for OARs were 60.2 Gy (IQR 57.9–67.5 Gy) for the rectum, 64.9 Gy (IQR 59.9–70.4 Gy) for the bladder, 61.6 Gy (IQR 58.5–69.1 Gy) for the sigmoid colon, and 63.4 Gy (IQR 59.5–67.1 Gy) for small intestine.

No Grade 3 or above acute or late toxicity was reported. The incidences of Garde 2 acute toxicity were 9.8% (n = 5) for rectal toxicity, 17.6% (n = 9) for hematologic toxicity, and 13.7% (n = 7) for urinary toxicity. The incidences of Garde 2 late toxicity were 9.8% (n = 4) for rectal toxicity, 7.8% (n = 4) for hematologic toxicity, and 0% for urinary toxicity. No Grade 2 acute or late vaginal toxicity was reported. Table 2. summarizes acute and late toxicity grouped by radiotherapy modalities (IGBT alone vs. IGBT + EBRT). No significant difference in Grade 2 or above acute or late toxicity was observed between IGBT alone and IGBT + EBRT groups. Chemotherapy increases the risk of Grade 2 or above acute or late hematological toxicity (χ² = 8.718, P = 0.007, OR = 12.67).

Table 2 Summary of dosimetric parameters and toxicities for radiotherapy

Survival outcomes

The median follow-up time was 58.0 months (IQR, 37.0–69.0). The rate of CR was 92.2% (n = 47), while the remaining 4 patients had post-treatment residual tumors, all of whom were high-risk patients and died within 1 year. The cumulative rates of locoregional failure and distant failure were 19.6% (n = 10) and 7.8% (n = 4), respectively. Among the locoregional failures, 4 patients had residual tumors after radiotherapy, and the remaining 6 patients experienced recurrence after a median follow-up of 22.5 months, with 5 cases being uterine recurrence and 1 case involving simultaneous uterine and pelvic lymph node recurrence. Regarding distant failures, 4 cases experienced distant metastasis after a median follow-up of 13.5 months, with the most common sites being lung, liver, and bone. As shown in Table 3., the 2-year, 3-year, and 5-year estimates of LRC were 85.8%, 80.4%, and 76.9%, respectively. The 2-year, 3-year, and 5-year estimates of TTP were 79.4%, 74.4%, and 71.2%, respectively.

20 patients (39.2%) died during follow-up, of which 10 patients (19.6%) had cancer-related deaths and 10 patients (19.6%) had comorbidity-related deaths. The mean survival times for cancer-related deaths and comorbidity-related deaths were 17.4 months and 32.5 months, with no significant difference between the two groups (P = 0.131). As shown in Table 3., the estimated OS at 2, 3, and 5 years were 78.3%, 71.9%, and 59.4%, respectively. The estimated CSS at 2, 3, and 5 years were 85.2%, 80.6%, and 77.0%, respectively. The LRC, TTP, OS and CSS curve of total patients are shown in Fig. 2A-D.

Risk classification

The 2-year, 3-year, and 5-year LRC, TTP, OS, and CSS for the low-risk and high-risk groups are shown in Table 3., and the corresponding survival curves are depicted in Fig. 2E-H. In terms of LRC, TTP, and CSS, low-risk group had significantly better survival outcomes compared to the high-risk group. The unadjusted HRs were 5.46 for LRC (95% CI 1.56–19.15), 7.69 for TTP (95% CI 2.55–23.14), and 5.86 for CSS (95% CI 1.66–20.71), with P-values all reaching statistical significance (P = 0.014 for LRC, P = 0.001 for TTP, and P = 0.010 for CSS). As for OS, there was a notable difference in survival rates between the low-risk and high-risk groups at the 2-year (92.3% vs. 63.3.6%) and 3-year (84.6% vs. 58.5%) time points, although the difference diminished at the 5-year time point (65.3% vs. 58.5%). Figure 2E. shows that the OS curves of the two groups overlap after approximately 65 months., and the final P value did not reach statistical significance (P = 0.090).

As shown in Table 1., there were significant differences between the low-risk and high-risk groups regarding age, number of comorbidity, radiotherapy mode, and chemotherapy (P < 0.1). After adjusting for these four factors using Cox proportional hazards regression models, the low-risk group continues to exhibit statistically significant superiority in terms of LRC, TTP, and CSS compared to the high-risk group (adjusted HR = 6.10, 95% CI: 1.18–31.45, P = 0.031 for LRC; adjusted HR = 8.07, 95% CI: 1.64–39.68, P = 0.010 for TTP; adjusted HR = 6.29, 95% CI: 1.19–33.10, P = 0.030 for CSS). In conclusion, the risk classification proposed in this study is an independent prognostic factor for IOEC patients receiving IGBT ± EBRT in terms of TTP, LRC, and CSS.

Table 3 Survival outcomes for low-risk and high-risk groups

Fig. 2

Survival curves for all patients and different risk groups. (A–D) LRC, TTP OS, and CSS curve for all patients. (E–H) LRC, TTP OS, and CSS curve for low-risk vs. high-risk groups

Discussion

Although the current standard care for endometrial cancer is staging surgery combined with adjuvant therapy determined by pathologic risk factors, an increasing number of EC patients are unable to tolerate surgery due to advanced age, obesity, multiple medical comorbidities, and/or extensive tumor lesions. For this special subgroup of EC patients, definitive radiotherapy (IGBT ± EBRT) can be used as an alternative treatment option for surgery. Although there has been a large body of literature confirming the apparent efficacy and acceptable toxicity of definitive radiotherapy in patients with IOEC [7, 14,15,16], the advantages of 3D IGBT for endometrial cancer and the risk factors that affect the prognosis of the treatment are still unclear. In this paper, retrospectively analyzing 51 IOEC patients from a single medical center in China, we demonstrated that IGBT ± EBRT has satisfactory disease control and acceptable toxicity reactions as definitive treatment for patients with IOEC, providing a risk classification based on FIGO 2009 staging and biopsy pathology to distinguish patients with varying treatment prognosis.

Previous researches on definitive radiotherapy for IOEC have predominantly focused on early-stage (Stage I and II) patients, achieving 2-year OS and LRC rates exceeding 85% and 90%, respectively [4, 17,18,19], which is obviously superior to the results observed in this study. However, it is worth noting that this study included a considerable proportion of patients with locally advanced disease or pre-existing distant metastasis before treatment (Stage III and IV patients accounted for 32.1%). The survival outcomes in this study are similar to those in previous studies with a higher proportion of Stage III and IV patients. For instance, S. Espenel et al. reported 27 cases of IOEC patients treated with IGBT plus EBRT (of which 17 cases were Stage III and IV, accounting for 63.0%) [20]. The median follow-up time was 36.5 months, with cumulative rates of local, pelvic, and distant failures at 19%, 7%, and 26%, respectively. The 5-year OS and 5-year DFS were 63% and 49.7%, respectively. Similarly, S. Mutyala et al. published a single-center retrospective study that included 32 patients (with Stage III and IV patients comprising 19%), indicating a 2-year LRC of 83% [21].

Notably, in this special and frail IOEC population, comorbidity-related deaths account for a non-negligible proportion of overall deaths. In this study, 10 patients died of pre-existing medical comorbidities such as cardiovascular accidents, accounting for half of the 20 patients with all-cause death. Previous studies have also highlighted the high risk of comorbidity-related deaths in IOEC patients [20, 21]. For instance, Podzielinski et al. reported a hazard ratio of 3.4 (95% CI = 1.4–9.4, P = 0.003) for deaths from other causes compared to cancer-specific deaths [22]. Furthermore, we observed that patients who died from comorbidity-related causes had a longer mean survival time compared to those who died from cancer-related causes (32.5 vs. 17.4 months, p = 0.131 in this study). This suggests that the primary sources of death risk vary in different phases following tumor treatment, with cancer-related deaths predominating in the early phase and comorbidity-related deaths prominenting in the later phase. The convergence of the OS curves for the high-risk and low-risk groups beyond the 5-year follow-up, as illustrated in Fig. 2E, also supports the hypothesis mentioned above. The continual deaths due to severe comorbidities among low-risk patients lead to a sustained decline in their OS curve. Therefore, we recommend that IOEC patients undergoing definitive radiotherapy should not only closely monitor endometrial carcinoma disease progression but also pay equal attention to the control and management of comorbidities. Especially for patients with 5-year tumor progression-free, the medical focus should shift more to the management of comorbidities.

This study demonstrated that despite multiple medical comorbidities and poor general health in IOEC patients, toxicity reactions after IGBT ± EBRT remain very rare and mild, with no Grade 3 or above acute or late toxicity reported. In a systematic review comparing the difference in toxicity between 2D and 3D brachytherapy, the incidences of Grade 3 or above acute or late toxicities were 7.2% for 2D technique and 1.4% for 3D technique [21]. The excellent performance of 3D IGBT in reducing toxicity is attributed to the ability of 3D image information to guide the correct placement of the applicator, as well as to accurately delineate adjacent OARs for proactive avoidance during treatment planning [4, 23, 24]. Regarding IGBT target delineation principles, previous literature has advocated prioritizing adherence to OAR dose constraints over coverage of CTV [25], and appropriate reduction of CTV dose can protect surrounding critical structures without compromising treatment efficacy and local control [18, 26]. In addition, this study found that combined chemotherapy significantly increased the incidence of Grade 2 or above acute or late hematologic toxicity (P = 0.007, OR = 12.67). This highlights the importance of a cautious evaluation of the overall health conditions of IOEC patients undergoing definitive radiotherapy, with particular attention to their tolerance to hematological toxicity, when devising comprehensive treatment strategies. In conclusion, the use of IGBT techniques, strict limitation of the OARs dose, and careful combination of chemotherapy are key measures to reduce the toxicity of definitive radiotherapy in IOEC patients.

The current commonly used risk classification for predicting prognosis and guiding treatment in endometrial cancer patients is mainly based on surgical pathology, which is unavailable in patients with IOEC. Therefore, we propose a new risk classification specific to IOEC patients according to long-term clinical experience and previous literature [26, 27], whose classification criteria are based on the FIGO 2009 staging determined by imaging, histological type and grade provided by biopsy pathology, aiming to predict the prognosis of IOEC patients undergoing definitive radiotherapy. Using the risk classification proposed in this paper to evaluate the prognosis of IOEC patients treated with definitive radiotherapy, there were significant differences in tumor control (LRC, TTP, CSS) between the low-risk and high-risk groups, but no significant difference in overall survival benefit (OS). Considering the elevated risk of comorbidity-related deaths in IOEC patients, it is hypothesized that the main reason for this is that the risk classification presented in this study only takes into account endometrial cancer-related factors and does not incorporate an assessment of comorbidity severity. For low-risk patients, receiving IGBT ± EBRT can achieve satisfactory disease control and survival benefits (with 5-year LRC, TTP, and CSS rates of 90.8%, 90.8%, and 90.2%, respectively), which significantly outperforms other available treatment options such as EBRT alone [28] or hormone therapy [29,30,31]. However, a large proportion of IOEC patients still do not receive standardized definitive radiotherapy, resulting in a poor prognosis [31, 32]. Thus, it is crucial to improve the utilization rate of definitive radiotherapy, and early referral of IOEC patients to high-volume centers is advantageous for optimizing their treatment management [5]. High-risk group patients also exhibit favorable prognosis; however, there is still a need to further exploration of improved treatment modalities to prolong survival, such as the combination of novel systemic therapeutic agents, including immunotherapy [5].

This study is a cohort research based on the Asian population, establishing the first risk classification to predict the prognosis of IOEC patients undergoing definitive radiotherapy. It also includes a considerable proportion of Stage III and IV patients to offer more generalizable data. In terms of limitations, it is difficult to avoid selection bias in a single-center retrospective study. Only 51 patients were included due to the rarity of IOEC patients, so the statistical analysis, especially the multifactorial analysis, was limited by the sample size, and therefore the results of the prognostic analysis need to be interpreted with caution. This study’s survival and disease control rates are comparable to previous literature, but the reported toxicity is lower. There is a possibility that the toxicity was underestimated because toxicity descriptions in some medical records were not standardized, which made the retrospective grading according to the CTCAE biased. In future studies, developing a risk classification model that simultaneously considers both endometrial cancer and comorbidities is recommended. This model could attempt to include an evaluation of comorbidity severity, such as the age-adjusted Charlson Comorbidity Index (aCCI), to effectively differentiate OS disparities among different risk groups. Furthermore, due to the significant competition between comorbidity-related and tumor-related deaths in IOEC patients, exploring risk-competition models as an alternative to conventional Cox proportional hazards regression models may enhance the accuracy of statistical findings.

Conclusion

For endometrial carcinoma patients who are unable to tolerate surgery due to medical comorbidities or extensive tumor invasion, IGBT ± EBRT has shown significant efficacy in improving disease control and prolonging survival without causing severe toxicity, thus it should be considered a preferred definitive treatment. The risk classification based on FIGO 2009 staging and biopsy pathology is of great value in predicting the prognosis of IOEC patients undergoing definitive radiotherapy, which serves as an independent prognostic factor for TTP, LRC, and CSS. It is worth noting that the IOEC patient population has a high risk of comorbidity-related death. The management of underlying diseases should be strengthened along with tumor control. Attempts can be made to incorporate the assessment of comorbidity severity into the risk classification to improve the prediction of OS.