Impact of resection for ovarian metastases from colorectal cancer and clinicopathologic analysis: A multicenter retrospective study in Japan

Abstract Aim The aim of this study was to clarify the significance of resection of ovarian metastases from colorectal cancer and to identify the clinicopathologic characteristics. Methods In this multicenter retrospective study, we evaluated data on ovarian metastases from colorectal cancer obtained from patients at 20 centers in Japan between 2000 and 2014. We examined the impact of resection on the prognosis of patients with ovarian metastases and examined prognostic factors. Results The study included 296 patients with ovarian metastasis. The 3‐y overall survival rate was 68.6% for solitary ovarian metastases. In all cases of this cohort, the 3‐y overall survival rates after curative resection, noncurative resection, and nonresection were 65.9%, 31.8%, and 6.1%, respectively (curative resection vs noncurative resection [P < 0.01] and noncurative resection vs nonresection [P < 0.01]). In the multivariate analysis of prognostic factors, tumor size of ovarian metastasis (P < 0.01), bilateral ovarian metastasis (P = 0.01), peritoneal metastasis (P < 0.01), pulmonary metastasis (P = 0.04), liver metastasis (P < 0.01), and remnant of ovarian metastasis (P < 0.01) were statistically significantly different. Conclusion The prognosis after curative resection for solitary ovarian metastases was shown to be relatively favorable as Stage IV colorectal cancer. Resection of ovarian metastases, not only curative resection but also noncurative resection, confers a survival benefit. Prognostic factors were large ovarian metastases, bilateral ovarian metastases, the presence of extraovarian metastases, and remnant ovarian metastases.


| INTRODUC TI ON
According to Global Cancer Statistics 2020, the number of new cases of colorectal cancer (CRC) worldwide is 1.88 million, ranking third after breast cancer and lung cancer.In addition, the number of deaths from CRC is 920,000, ranking second after lung cancer. 1 Approximately 1.1%-4.2% of women with CRC are diagnosed with ovarian metastasis, 2,3 and median survival for these women ranges from 19 to 27 mo. 4,5 recent years, chemotherapy including molecular-targeted therapy has evolved, improving the survival time for metastatic CRC.
However, ovarian metastases are less responsive to chemotherapy than other sites, and the ovary is characterized as a "sanctuary for metastases." 6Surgical resection should always be considered in cases of ovarian metastases, even in the presence of extraovarian metastases.
The pathway leading from CRC to ovarian metastasis remains unclear.Several theoretical routes that may involve direct invasion, peritoneal dissemination, 7 hematogenous metastasis, or lymphatic metastasis. 8,9Ovarian metastasis is currently classified as distant metastasis (M) according to both the Japanese Classification of Colorectal, Appendiceal, and Anal Carcinoma (JCCRC), 10 and TNM classification systems, although the Japanese classification system previously classified it as peritoneal metastasis.
Previous reports have analyzed the prognosis, prognostic factor, treatment, and characteristics by small sample sizes at a single institution.However, cases with ovarian metastases are often accompanied by other distant metastases, and robust data on clinicopathologic characteristics, prognosis, and the effect of treatment are lacking.Consequently, no consensus on the management of ovarian metastases from CRC worldwide exists.The significance of surgery for ovarian metastases and treatment strategies, including prophylactic and palliative ovarian resection, are controversial.
This study was conducted as a multicenter and retrospective study to clarify the clinicopathological features, prognosis, and significance of ovarian resection for ovarian metastasis from CRC. participating hospital.The protocol of this study conforms to the provisions of the Declaration of Helsinki.Each patient provided informed consent for the use of their data, in accordance with the institutional regulations of the respective participating institution.
The protocol of this study was registered at the University Hospital Medical Information Network Clinical Trial Registry (UMIN000029371).

| Patients and methods
The study included patients diagnosed with ovarian metastasis from CRC between 2000 and 2014 from the 20 Japanese institutions.
Primary CRC was histologically diagnosed as adenocarcinoma.The diagnosis of ovarian metastasis was based on histological findings in cases that involved ovarian resection, or imaging findings (e.g.computed tomography, positron emission tomography, and magnetic resonance imaging) in cases that did not resect ovarian metastasis.
Cases were excluded if they involved oophorectomy for other diseases, the previous treatment of other pelvic malignancies, or appendix cancer.
Patient data were collected retrospectively from medical records containing the following information: age, primary tumor site, histology, TNM classification, site of distant metastasis, vascular invasion, lymphovascular invasion, preoperative serum carcinoembryonic antigen level, preoperative serum carbohydrate antigen 19-9 level, resection of primary tumor, curability according to JCCRC, 10

| Definition
The overall survival (OS) period was defined as the survival period after surgery in cases with ovarian resection and the survival period from the date of imaging diagnosis in cases without ovarian resection.Synchronous metastases were defined as those diagnosed at the same time as the primary tumor, and metachronous metastases were defined as those diagnosed after the primary tumor had been In the resection of ovarian metastases, curative resection is defined as R0.Noncurative resection describes surgery for the symptomatic management of enlarged ovaries or cytoreduction, even though distant metastases at other sites may remain.Nonresection referred to ovarian metastases that had not been resected.Patients with nonresected ovaries were treated with chemotherapy, whereas those with remnant ovarian metastases were not administered any treatment.Remnant ovarian metastases, which were defined as unresected ovarian metastases, included cases in the nonresection group.

| Statistical analysis
Continuous data are shown as median (range), and categorical data as number (%).Any differences between continuous and categorical variables were determined using the Mann-Whitney U test and the chi-square test or Fisher's exact test, respectively.The OS interval was calculated based on the duration from the diagnosis of an ovarian metastasis (whether synchronous or metachronous) to the date of death due to any cause.Survival curves were estimated using the Kaplan-Meier method and compared using the log-rank test.Variables deemed to be significant in the univariate analyses were included in a Cox proportional hazards model, and the results were expressed as hazard ratios (HRs) and 95% confidence intervals (CIs).P values <0.05 were significantly different.All statistical analyses were conducted using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria, v. 2.13.0).

| Patient characteristics
Between 2000 and 2014, a total of 20,841 female patients underwent surgery for CRC at the 20 institutions.Among these patients, 296 cases of ovarian metastasis from CRC were identified, which indicates an estimated incidence of 1.4%.A total of 116 cases of synchronous ovarian metastasis and 180 cases of metachronous ovarian metastasis exists.
The characteristics of the primary tumors of the patients who participated in this study are presented in

| Oncological outcomes
The median duration of follow-up of the study population was 36.6 mo (range: 1-201 mo).Table 4 shows the recurrence sites in cases of curative resection (n = 121) of the metastases is commonly found in the peritoneum (47 cases, 38.8%), followed by the liver (25 cases, 20.6%) and the lung (24 cases, 19.8%).Among the 28 patients who underwent unilateral oophorectomy, recurrence occurred in the contralateral ovary in five cases (17.9%).

| Prognostic factors
3][14] In this study, the noncurative group had a better prognosis than those in the nonresection group.Notably, patient backgrounds of the two groups revealed no significant difference in the frequency of other distant metastases, which is a prognostic factor.The nonresection group included more patients with older age, metachronous ovarian metastases, and ovarian metastases <10 cm in size.Bias in patient background may possibly be involved in whether the patient is in a condition to undergo oophorectomy.
Verifying whether palliative surgery improves prognosis over chemotherapy is difficult.While ovarian metastasis is considered a sanctuary for metastasis, oophorectomy for ovaries resistant during chemotherapy was reported to allow continuation of chemotherapy and prolong the prognosis. 15,16In this study cohort, subgroup analysis of prognosis for the 2000-2007 and 2008-2014 periods revealed no change in the prognosis of patients with ovarian metastasis (Figure S1).Despite improvements in systemic chemotherapy, no improvement in prognosis was observed, suggesting that ovarian metastasis is a sanctuary for metastasis.Palliative resection should be considered for ovarian metastases during chemotherapy for unresectable metastatic CRC.
The decision to perform bilateral oophorectomy in cases of unilateral ovarian metastases is an important question for the surgeon.
In this study, 17.9% of patients who underwent unilateral oophorectomy had recurrence in the other ovary.In contrast, Zhou et al.
reported recurrence in the contralateral ovary in 44% of cases in which unilateral oophorectomy was performed. 14Mori et al. reported that 56% of patients with preoperative diagnosis of unilateral ovarian metastasis had bilateral ovarian metastasis. 17The bilateral ovarian resection in these situations may be beneficial, considering that ovarian metastases are less sensitive to chemotherapy and may be affected by potential ovarian metastases.The American Society of Colon and Rectal Surgeons Clinical Practice Guidelines state that bilateral oophorectomy should typically be performed even if one ovary appears grossly normal for patients with suspected or known metastatic disease involving an ovary. 18 the TNM classification, ovarian metastasis is defined as a distant metastasis; however, ovarian metastasis was previously recognized as a phenotype of peritoneal dissemination in Japan and was classified as peritoneal metastasis in the past JCCRC.The metastatic pathways that lead to ovarian tumors remain unclear.Fujiwara et al.
suggested that hematologic or lymphatic spread of malignant cells to the ovaries was the most likely pathway, based on the pathological examination of metastatic ovarian tumors. 9A review of the Swedish Cancer Registry stated that ovarian metastases are more frequently associated with peritoneal metastases, logically, as the peritoneum and ovaries are often considered a continuum, and ovarian metastases are rarely seen in rectal cancers due to anatomic factors such as extraperitoneal structures. 7 this study the discovered characteristics of ovarian metastasis were suggestive of an association with peritoneal metasta- of these patients had peritoneal recurrence. 19Thus, the frequency of peritoneal metastatic recurrence after the resection of ovarian metastases is comparable to that after the resection of peritoneal metastases.Therefore, from a clinicopathological perspective, we suggest that CRC ovarian metastasis appears to be strongly associated with peritoneal metastasis.
If ovarian metastasis were to be classified as peritoneal metastasis, the following treatment outcomes of peritoneal metastasis would be compared and verified.The JCCRC defines peritoneal metastasis as follows: P0, no peritoneal metastasis; P1, metastasis localized to the adjacent peritoneum; P2, limited metastasis to the distant peritoneum; and P3, diffuse metastasis to the distant peritoneum.According to the system, higher numbers are associated with poorer prognoses, as indicated by the Study Group for Peritoneal Metastasis from Colorectal Cancer from the JSCCR.Peritoneal metastasis after R0 surgery has been associated with a 3-y OS rate of 42.0% and a 5-y OS rate of 36.2%, with 3-y OS rates of 45.7% for P1 disease, 39.5% for P2 disease, and 25.0% for P3 disease. 19e 5-y OS rates have been reported as 32.4% for R0 resection and 4.7% for R1/R2 resection. 20In the present study, curative resection of ovarian metastasis was associated with a 3-y OS rate of 65.9%, which was better than the corresponding rate determined for patients with P1 disease after R0 resection.Therefore, assuming that ovarian metastasis is classified as peritoneal metastasis, we consider

AUTH O R CO NTR I B UTI O N S
All authors substantially contributed to the conception and design and acquisition of the study.HK, KM, and YK performed statistical analysis and data interpretation.HK and YK drafted the article, and all authors critically revised it for important intellectual content.All authors gave final approval for this version to be published.

ACK N OWLED G M ENTS
We thank to Tsuyoshi Konishi MD, PhD, Keiichi Takahashi MD, PhD, Koutaro Maeda MD, PhD, and Shigeki Yamaguchi MD, PhD for the design and conception and acquisition of this study.We thank Editage (www.editage.jp) for editing a draft of the article.

FU N D I N G I N FO R M ATI O N
This work was supported by the Japanese Society for Cancer of Colon and Rectum.

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declare no conflicts of interest for this article.Kay Uehara MD, PhD is an editorial member of the Annals of Gastroenterological Surgery.

E TH I C S S TATEM ENT
Approval of the research protocol: The study protocol was approved by the Institutional Review Boards of all participating hospitals.
Informed Consent: N/A.
size of ovarian tumor, timing of ovarian metastasis, unilateral or bilateral ovarian metastasis, rupture of the ovarian tumor, treatment for the ovarian tumor, unilateral or bilateral oophorectomy, intention to undergo surgical treatment, postoperative chemotherapy, recurrence after resection of ovarian metastasis, and survival interval.The primary outcome assessed was survival after diagnosis of the ovarian metastasis, which was evaluated according to curative resection, noncurative resection, or nonresection of the ovarian metastasis.The secondary outcomes assessed included prognostic factors for ovarian metastasis in CRC, the clinicopathological characteristics of ovarian metastasis from CRC, and survival of solitary ovarian metastases.
resected.The definition of curability for primary tumors is defined as follows in accordance with the JCCRC: Curability A (CurA): No distant metastasis (M0), and no residual tumor at both proximal/distal and radial margins; Curability B (CurB): Not corresponding to Curability A or C; Curability C (CurC): Macroscopic residual tumor.

TA B L E 1 Characteristics of primary tumor Variable All cases n = 296 Synchronous OM n = 116
had peritoneal metastasis.Meanwhile, 34 patients (18.9%) had peritoneal metastasis in cases of metachronous ovarian metastasis.In the 73 cases of synchronous ovarian metastases (62.9%), the primary tumor was noncurative resected.The characteristics of ovarian metastases from CRC are presented in Table2.Bilateral metastases were found in 112 patients (37.8%) and unilateral metastases in 184 patients (62.2%).The median tumor diameter of ovarian metastases was 10 cm.The median disease-free interval duration after primary resection of metachronous ovarian metastasis was 13.1 mo.Solitary ovarian metastasis occurred only in 22 cases (19.0%) of synchronous ovarian metastasis, while 66 cases (36.7%) of metachronous ovarian metastasis occurred (P < 0.01).Bilateral oophorectomy was performed in 191 patients (64.5%).Resection of the ovarian metastasis was performed in 252 patients (85.1%), with surgeries classified as curative resection in 121 cases (40.9%) and noncurative resection in 128 cases (43.2%).

Variable All cases n = 296 Synchronous OM n = 116
TA B L E 1 (Continued)

Variable All cases n = 296 Synchronous OM n = 116
Patient characteristics categorized by the curability of surgical treatment [11][12][13][14]cteristics of ovary metastasisprolong the prognosis, prognostic factors, and characteristics of ovarian metastases.This retrospective study examined ovarian metastases from CRC, which is the largest number of cases in previous reports, excluding the national registry study.3OvarianmetastasisfromCRC,which occurs relatively rarely, is often a complex condition with variable timing of development and other distant metastases; hence, investigating many cases is necessary.In this study, solitary ovarian metastasis, whether synchronous or metachronous ovarian metastasis, is expected to have a prolonged prognosis.According to the results of this cohort, the 3-y OS of solitary ovarian metastasis was 68.6%, which is a relatively promising prognosis among Stage IV CRC patients.Previous studies examining prognostic factors have reported extraovarian metastasis as an independent prognostic factor.[11][12][13][14]Becausemost ovarian F I G U R E 1 Overall survival of solitary ovarian metastasis.Overall survival curves are shown for patients with synchronous OM and metachronous OM.Black line: Synchronous OM, Red line: Metachronous OM.OM: ovary metastasis.

Variable All cases n = 296 Curative resection (CR) n = 121 Noncurative resection (NCR) n = 128 Nonresection (NR) n = 47 CR vs NCR P value NCR vs NR P value CR vs NR P
Recurrence site after curative resection for ovarian metastasis TA B L E 4icopathologic characteristics of ovarian metastases from a larger number of CRC cases in this multicenter retrospective study.The results showed that the prognosis of solitary ovarian metastases is relatively favorable in Stage IV CRC.Furthermore, resection of ovarian metastases in both curative and noncurative is beneficial for survival.Ovarian metastasis from CRC was suggested to be associated with peritoneal metastasis.
Prognostic factors in patients with ovarian metastasis Registry and the Registration No. of the study/trial: The clinical trial registration number is UMIN000029371.Animal Studies: N/A.TA B L E 5Abbreviations: CA 19-9, carbohydrate antigen 19-9; CEA, carcinoembryonic antigen; CI, confidence interval; LN, lymph node; OM, ovarian metastasis; OR, odds ratio.