Adjuvant Pelvic Radiotherapy in Ovarian Cancer: Outcomes By Risk Factors

Background: Radiotherapy (RT) is not the current standard treatment option for ovarian cancer (OC) patients. The present study aims to explore the risk factors of adjuvant pelvic RT for patients with OC and to investigate prognostic factors associated with longer progression free survival (PFS) and overall survival (OS) period. Methods: From April 2004 to September 2020, OC patients administered with RT were collected retrospectively at our institution and those who received pelvic RT were selected for further analysis. Kaplan-Meier method was employed to estimate the PFS, OS and local control (LC) rate. Univariate and multivariate cox regression model were established to identify potential beneciaries of pelvic RT. Results: Overall, a total of 89 OC patients underwent RT were identied, 70 of whom were treated with adjuvant pelvic RT and were eligible for nal analysis. Median follow-up was 51.0 months. The estimated 3-year OS rate was 48.1% and 3-year PFS rate was 26.8%, respectively. CACTE grade 4 to 5 toxicities were observed in 2 patients. Multivariate cox model illustrated that high-grade serous carcinoma (HGSC), advanced stage disease (FIGO stage III & IV), receiving pelvic RT during multiple relapses as well as no previous chemotherapy (CT) were independent risk factors for PFS. Additionally, patients with 0-1 risk factor had a longer PFS time than those with 2-4 risk factors. Conclusions: Pelvic RT was well tolerated and had favorable ecacy in OC patients with low risk factor (0-1), which might confer prolonged survival. However, further investigations are needed.


Background
Ovarian cancer (OC), a common gynecological malignancy, accounts for the fth leading cause of cancer-related mortality in women (1). The established cornerstone treatment for OC patients consists of aggressive cytoreductive surgery and platinum based chemotherapy (CT) (2). Although most patients demonstrate an optimal response to the gold treatment and enter into clinical remission, over 70% of them will experience relapse during the course of disease, leading to a 5-year survival rate of approximately 30% (3).
Historically, radiotherapy (RT) served a dominant role in the postoperative adjuvant setting for patients with early stage or minimal residual disease (4). Over years, systematic platinum-based CT, obtaining higher response rates, were introduced to clinical practice and consequently, in the 1980s, RT were used sparingly for OC patients due to the con icting clinical ndings as well as severe toxic events including myelosuppression and gastrointestinal intolerance (5).
More recently, given the improvement of RT delivery techniques such as intensity-modulated radiotherapy (IMRT) and the exhibited platinum-resistance, the role of RT in speci c clinical scenarios ought to be rede ned. A prospective research gured out that whole abdomen irradiation delivered with IMRT technique could be performed with acceptable toxicity for patients receiving radical cytoreductive surgery and platinum-based adjuvant CT (6). Along with the gratifying results in terms of overall survival (OS) and progression free survival (PFS) observed in a subsequent trial, whole abdomen irradiation might offer a potential therapeutic option for advanced stage OC patients (7). However, limited literatures shed light on the identi cation of patients who would in particular stand to bene t from RT.
In the present study, we retrospectively assessed the e cacy and toxicity of pelvic RT for OC patient population in the adjuvant setting, aiming to de ne the potential bene ciaries of pelvic RT, which would be of utmost importance to procure a longer regional control.

Study population
We conducted a retrospective study, reviewing the data of 89 OC patients that received RT from April 2004 to September 2020 in the Department of Radiation and Medical Oncology of Zhongnan hospital of Wuhan University. Clinical information was derived from the electronic medical records system of our institution. We obtained permission from the Ethics Committee of Zhongnan hospital of Wuhan University for data acquisition and analysis, and all patients enrolled gave a verbal or written informed consent. Inclusion criteria were as follows: (a) histologically diagnosed ovary malignancy; (b) received pelvic RT either at the time of initial treatment or recurrence; (c) follow up for ≥ 3 months. A total of 70 OC patients were nally included in the analysis.
According to the 2014 WHO guidelines, we further categorized grade II, III and IV serous carcinomas and reclassi ed grade III/IV endometrioid into high-grade serous carcinoma (HGSC). Tumors with other histological types were considered as non-high-grade serous carcinoma (NHGSC). Initial stage of disease was assigned using the International Federation of Gynecology and Obstetrics (FIGO) criteria. Platinumsensitive and platinum-resistant were de ned as a progression/relapse more than or within 6 months from the last dose platinum-based CT, respectively. Clinical characteristics are displayed in Table 1.

RT policy
In cases with external beam radiotherapy (EBRT), patients were placed in a supine position with arms elevated overhead. The clinical target volume (CTV) encompassed the entire pelvic cavity and draining lymph node regions, ranging from the 5th lumbar to the bottom of symphysis pubis. The planning target volume (PTV) was de ned as extending the CTV by 7-10 mm in all dimensions. Organs at risk (OARs) included small intestine, colon, bladder, femoral head and neck, and spinal cord. Dose was prescribed to cover 95% of PTV with 100% isodose line and hot spot regions were localized in CTV. OAR dose criteria were taken from the published RTOG 0418 protocol. Radiation dose planning was performed on the CMS XiO (Varian, USA) treatment planning system (TPS) utilizing a superposition/convolution algorithm. IMRT plans were performed using multiple noncoplanar photon beams. Following EBRT, 5 patients underwent additional intracavity brachytherapy by a high-dose-rate afterloading system using iridium-192 source which was performed once or twice per week, with a cumulative dose ranging from 12 Gy to 33 Gy. RT characteristics are displayed in Supplementary table 1.

Post-treatment follow up
Follow-up data was obtained from clinical examinations and/or telephone survey. Clinical examination items including routinely physical examination, chest computed tomography, abdominal computed tomography, pelvic magnetic resonance imaging, and serum tumor markers were performed every 3 months for the rst 2 years, semiannually for the next 3 years and annually thereafter. Patients were observed until death or loss to follow-up. OS was de ned as the date elapsed between RT initiation and death from any cause or the last follow-up visit. PFS was de ned as time from start of RT to the rst observation of progression, relapse or death, whichever occurred rst. Local control (LC) was de ned as no new tumors within the previous radiation eld.
Treatment-related complications were also recorded during follow-up and graded on the basis of National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE), Version 4.0.
We employed Kaplan-Meier method was employed to compute OS, PFS, LC rates with 95% con dence intervals (CIs). Univariate cox proportional hazards regression model was established to gure out the prognostic factors. Variables with statistical signi cance (p < 0.1) or clinical signi cance were selected for the multivariate model. A two-sided p-value < 0.05 was considered to re ecting statistical signi cance.

Population
The cohort of 70 OC patients treated with pelvic RT identi ed in the current study had a median age of 54 years (34-77 years

Univariate and multivariate analysis
Prognostic factors analysis based on univariate regression model was displayed in Table 2. The result indicated that histology, initial FIGO stage, surgery before RT, and history of relapse exerted a most pronounced in uence on PFS (p < 0.01). Along similar lines, platinum-sensitivity was also strongly correlated with PFS (p < 0.05). Age, chemotherapy cycle before RT and family history of cancer were not signi cant factors for PFS. Initial FIGO stage and surgery before RT were found to in uence OS (p < 0.05), while no effect of other factors on OS was seen. In terms of LC, none of the predictors in the model had a signi cant effect on LC. Given that chemotherapy cycle before RT was also associated with response to RT and survival outcome (8, 9), it was also included in the multivariate model other than variables with statistical signi cance (p < 0.1). As shown in Fig. 2A, it was noticeable that patients dispensed with pelvic RT during initial treatment and rst relapse displayed better PFS (p < 0.05) compared with those treated with pelvic RT when experiencing multiple relapses. Meanwhile, NHGSC and more than 6 cycles of CT before RT were related with a longer PFS (p < 0.05  Fig. 3, PFS time in high risk group was signi cantly shorter than in low risk group, indicating that more aggressive and effective treatment regimens should be taken into practice to achieve a better local control for high risk patients. Adverse events CTCAE, Version 4.0 grade 4 to 5 toxicities were observed in 2 patients (Table 3). One patients received palliative pelvic RT and died of severe intestinal obstruction after the completion of RT, which was mainly due to intestinal adhesion caused by surgery. Another patient developed grade 4 acute gastrointestinal toxicity and relieved by medication. A total of 29 patients (29/70, 41.4%) experienced grade 1 to 2 acute gastrointestinal toxicity and 14 (14/70, 20.0%) had grade 1 to 2 hematological toxicity. Only 4 patients displayed grade 3 hematological toxicity. Other adverse reactions related to RT included grade 1 or grade 2 genitourinary toxicity (n = 5) and grade 1 dermatitis (n = 1). Generally, pelvic RT was well-tolerated in most patients and could be performed as previously planned without interruption of RT related toxicity.

Discussion
In the current single-center study, 70 cases of OC patients administered with pelvic RT were retrospectively reviewed. Our data revealed that pelvic RT exhibited remarkable antitumor activity in OC patients without unpredictable adverse events. We reported a 2-year OS rate of 69.1% and a 2-year PFS rate of 34.8%, which corroborated with the ndings of previous investigations (10,11). The 3-year LC rate was 75.5%, supporting recent studies documenting that RT yielded satisfying local control outcome in OC patients (10,12,13). Besides, in agreement with other researches (14,15), survival after RT had no correlation with platinum sensitivity at the onset of RT, indicating that RT could serve as an effective modality in the treatment of OC patients irrespective of platinum status.
Univariate and multivariate cox regression models were constructed in an attempt to locate potential bene ciaries of pelvic RT. Concerning histology, we observed that NHGSC was a favorable prognostic factor. Rare histology variants, such as ovarian clear cell carcinoma (OCCC), was particularly chemoresistant with a clinical response rate to platinum-based CT of merely 18% (16, 17), and chances of curability of this subtype could be enhanced by adjuvant RT (18 Collectively, early and aggressive RT ought to be employed in the management of patients diagnosed with OCCC. In the current study, early stage disease (FIGO stage I & II) and more CT regimens before RT predicted a longer PFS rate and OS rate. It is a well-acknowledged fact that advanced stage OC patients recurred after a median interval of 18-24 months and had a poor 5-year OS rate of 31% for stage III, 13% for stage IV (14), respectively, compared with early stage disease. This nding is inconsistent with that of Robert Rome et al. who observed signi cant correlation between initial stage I & II and disease speci c death (19). However, prior investigations have shown that patients with stage III OC could still gain bene t from adjuvant intensity modulated whole-abdominal radiotherapy although more aggressive and accurate local management regimens are in urgent need (7). Regarding numbers of prior CT regimens, our nding is contrary to previous study which suggested that patients exposed to one prior CT regimens showed longer duration of survival than those administered with more prior CT regimens (15). The reason why numbers of CT cycles before RT was identi ed as a prognostic factor remains uncovered, and further mechanistic research should be undertaken to explore the underlying mechanism. It's highly likely that this result could be a statistical anomaly due to the limited number of patients enrolled in our retrospective study.
Furthermore, adjuvant RT after rst relapse could markedly extend PFS period, but the OS bene t had not been observed according to our result. Westhoff et al. also found a prolonged PFS in 10 patients received RT for the rst recurrence compared with those received it for subsequent recurrence (20). Additionally, we also reached the conclusion that adjuvant RT improved PFS period for OC patients during initial treatment. However, we didn't compare the toxicity and survival differences between post-operative adjuvant RT, adjuvant CT alone, and sequential chemoradiotherapy at initial treatment, which was in demand for in-depth investigations to con rm and validate these ndings. Thus, we are unable to comment on the superiority of adjuvant RT over adjuvant CT for initial treatment from the results obtained in this study.
Several limitations to this study need to be acknowledged. First, it was a single-center experience and the sample size was relatively small. Second, it is possible that the results were biased, given the nature of retrospective design. Selection biases might occur when gynecologic oncologists recommended adjuvant pelvic RT. Finally, patients' economic conditions and medical literacy would exert a large effect on treatment choice. Hence, de nitive large clinical trials are awaited regarding suitable OC patients for adjuvant pelvic RT.
In conclusion, pelvic RT might be a promising treatment option for OC patients, regardless of the platinum-sensitivity status. Patients with early stage disease, NHGSC, experiencing initial treatment and rst recurrence, receiving more CT cycles are good candidates for adjuvant pelvic RT. Further investigations are warranted to con rm and validate these factors in terms of choosing treatment for OC patients.

Declarations
Ethics approval and consent to participate All the procedures performed in studies involving human participants were in accordance with the ethical standards of the Ethics Committee of Zhongnan hospital of Wuhan University and with the 1975 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Consent for publication
Not applicable Availability of data and materials All data generated or analysed during this study are included in this published article and its supplementary information les.

Competing interests
The authors declare that they have no competing interests. Kaplan-Meier survival curves on OS, PFS, LC of ovary carcinoma patients received pelvic radiotherapy. OS, overall survival; PFS, progression free survival; LC, local control; CI, con dence interval; NR, not reached Figure 2 Forest plot based on the results of multivariate analysis of prognostic factors associated with (A) PFS and (B) OS, PFS, progression free survival; OS, overall survival; HR, hazard ratio; CI, con dence interval;