Cancer of the ovary, fallopian tube, and peritoneum: 2021 update.

JB, MR, SK, LK, and MF reviewed and updated the chapter on cancer of the ovary, fallopian tube, and peritoneum published in the 2018 Cancer Report.

In general, the prognosis of epithelial ovarian, fallopian, and peritoneal malignancies is independently affected by the following 1 , 52 , 53 : Stage of the cancer at diagnosis. Histologic type and grade. Maximum diameter of residual disease after cytoreductive surgery. Stage of the cancer at diagnosis. Histologic type and grade. Maximum diameter of residual disease after cytoreductive surgery. A thorough staging laparotomy is an important part of early management. If the preoperative suspicion is malignancy, a laparotomy should be performed. If there is no visible or palpable evidence of metastasis, the following should be performed for adequate staging 1 , 10 , 11 : Careful evaluation of all peritoneal surfaces. Retrieval of any peritoneal fluid or ascites. If there is none, washings of the peritoneal cavity should be performed. Infracolic omentectomy. Selective lymphadenectomy of the pelvic and para‐aortic lymph nodes, at least ipsilateral if the malignancy is unilateral. Biopsy or resection of any suspicious lesions, masses, or adhesions. Random peritoneal biopsies of normal surfaces, including from the undersurface of the right hemidiaphragm, bladder reflection, cul‐de‐sac, right and left paracolic recesses, and both pelvic sidewalls. Total abdominal hysterectomy and bilateral salpingo‐oophorectomy in most cases. Appendectomy for mucinous tumors if the appendix appears abnormal. Careful evaluation of all peritoneal surfaces. Retrieval of any peritoneal fluid or ascites. If there is none, washings of the peritoneal cavity should be performed. Infracolic omentectomy. Selective lymphadenectomy of the pelvic and para‐aortic lymph nodes, at least ipsilateral if the malignancy is unilateral. Biopsy or resection of any suspicious lesions, masses, or adhesions. Random peritoneal biopsies of normal surfaces, including from the undersurface of the right hemidiaphragm, bladder reflection, cul‐de‐sac, right and left paracolic recesses, and both pelvic sidewalls. Total abdominal hysterectomy and bilateral salpingo‐oophorectomy in most cases. Appendectomy for mucinous tumors if the appendix appears abnormal. Upon opening the abdominopelvic cavity, the peritoneal fluid should be sent for cytology. In the absence of ascites, irrigation should be performed and washings sent for cytology. The laparotomy should proceed with a detailed examination of the contents, including all of the peritoneal surfaces. In addition to the suspicious sites, biopsies from the peritoneal reflection of the bladder, the posterior cul‐de‐sac, both paracolic gutters, subdiaphragmatic surfaces, and both pelvic sidewalls should be taken. The primary tumor, if limited to the ovary, should be examined to look for capsular rupture. All obvious sites of tumor must be removed wherever possible in addition to total hysterectomy and bilateral salpingo‐oophorectomy. The omentum, pelvic, and para‐aortic lymph nodes should be removed for histologic examination. In younger women, fertility preservation may be desired. In these patients, conservative surgery, with preservation of the uterus and contralateral ovary, should be considered after informed consent. 47 Clinical judgment is important in the approach to a pelvic mass in the young, reproductive‐aged woman. If the suspicion is strong for malignancy, open laparotomy is generally indicated. Laparoscopy may be more appropriate if the suspicion is more for benign disease, where tumor markers (including hCG and AFP) are normal. A biopsy of any suspicious lesion can be performed and frozen section obtained in order to proceed expeditiously with definitive surgery. Ovaries and fallopian tubes should be evaluated as thoroughly as possible to establish the site of origin. If visible, the entire tube, particularly the distal portion, should be submitted for pathology and examined using the SEE‐FIM protocol. 33 Ovaries should be scrutinized for coexisting endometriotic cysts, adenofibromas, or other benign conditions that could serve as a nidus of tumor development. At least two‐thirds of patients with ovarian cancer present with Stage III or IV disease. This may affect the performance status and fitness for surgery. However, the most important prognostic indicator in patients with advanced stage ovarian cancer is the volume of residual disease after surgical debulking. Therefore, patients whose medical condition permits should generally undergo a primary laparotomy with total abdominal hysterectomy, bilateral salpingo‐oophorectomy, omentectomy, and maximal attempt at optimal cytoreduction. 1 , 52 , 53  This may necessitate bowel resection, and occasionally, partial or complete resection of other organs. Based on recent data from the randomized Lymphadenectomy in Ovarian Neoplasm (LION) trial, the removal of clinically negative lymph nodes during cytoreductive surgery does not increase the progression‐free or overall survival and should not be undertaken. 54 Level of Evidence A . In selected patients with cytologically proven Stage IIIC and IV disease who may not be good surgical candidates, 3–4 cycles of neoadjuvant chemotherapy (NACT) may be given initially, followed by interval debulking surgery and additional chemotherapy as demonstrated in the EROTC and CHORUS Trials. 55 , 56  These two randomized prospective trials showed that in selected patients, interval debulking surgery after neoadjuvant chemotherapy showed equivalent survival with less morbidity compared with primary cytoreductive surgery. NACT followed by interval debulking surgery may be particularly useful in patients with a poor performance status, significant medical comorbidities, visceral metastases, and those who have large pleural effusions and/or gross ascites. 57 , 58 In selected patients whose primary cytoreduction is considered suboptimal, particularly if a gynecologic oncologist did not perform the initial operation, interval debulking may be considered after 2–3 cycles of systemic chemotherapy. 1 , 55 , 56 , 59  Pathologic assessment for residual tumor following neoadjuvant therapy will enable an estimate of residual disease and pathological response. 60  There are recent data to indicate that patients who have a good pathological response have a better outcome. A histopathologic scoring system for measuring response to neoadjuvant chemotherapy has been developed and validated by Bohm et al. 61 who reported criteria for defining a chemotherapy response score (CRS) based on a three‐tier system. A CRS 3 (complete or near complete pathological response) was associated with a better prognosis. Recently, these results have been validated in an independent West Australian cohort. 62

Ovarian sarcomas are rare and occur primarily in postmenopausal patients. 161 , 189  Nevertheless, accurate diagnosis and differentiation from other types of primary ovarian cancer are important, as the prognosis is generally poor. There are two types of sarcoma. Malignant mixed Müllerian tumors (MMMTs) or ovarian carcinosarcomas, the more common of the two, are biphasic tumors composed of both carcinomatous and sarcomatous elements. 189 , 190  Most authors agree that most MMMTs are monoclonal in origin and should be thought of and managed as a high‐grade epithelial cancer. The sarcomatous component is derived from the carcinoma or from a stem cell that undergoes divergent differentiation. Thus, ovarian carcinosarcomas are best regarded as metaplastic carcinomas. Pure sarcomas are very rare and should be treated according to the specific histologic subtype. These rare sarcomas include fibrosarcomas, leiomyosarcomas, neurofibrosarcomas, rhabdomyosarcomas, chondrosarcomas, angiosarcomas, and liposarcomas. Their management is not discussed here. Patients with early stage MMMTs/ovarian carcinosarcomas have a better outcome than those with advanced stage disease, but the overall prognosis is poor. They should be managed similarly to high‐grade pelvic serous cancers. Their rarity prohibits any prospective randomized trials. The principles of surgical management of ovarian MMMTS/ovarian carcinosarcomas are the same as for high‐grade pelvic serous cancers. Following surgery, patients should receive platinum‐based chemotherapy. 161  The follow‐up schedule is as recommended for epithelial malignancies. Level of Evidence C .

Patients with epithelial ovarian cancers confined to the ovary or fallopian tube at initial diagnosis have a very good prognosis. 46 , 47 , 48 , 49  The symptoms are often very insidious and the duration of symptoms not very different between patients with early stage or advanced stage disease. 13 , 14  This may reflect the different biological behavior of the various histologic subtypes; for example, grade 1 serous, clear cell, mucinous, and endometrioid cancers are commonly early stage at presentation, whereas high‐grade serous cancers are most often Stage III because of early dissemination by a more aggressive cancer. Tumor markers such as human gonadotropin (hCG) and alpha‐fetoprotein (AFP) are mandatory to exclude germ cell tumors in younger patients with a pelvic mass or suspicious enlargement of an ovary. Approximately two‐thirds of all epithelial “ovarian” cancers are Stage III or Stage IV at diagnosis. Presenting symptoms include vague abdominal pain or discomfort, menstrual irregularities, dyspepsia, and other mild digestive disturbances, which may have been present for only a few weeks. 13 , 14 , 50 As the disease progresses, abdominal distention and discomfort from ascites generally worsen, and may be associated with respiratory symptoms from increased intra‐abdominal pressure or from the transudation of fluid into the pleural cavities. Abnormal vaginal bleeding is an uncommon symptom. Serous fallopian tube and peritoneal cancers present the same as ovarian cancer. Past analyses have been biased because many fallopian tube cancers have been presumed to arise in the ovaries. A detailed medical history must be taken to ascertain possible risk factors, history of other cancers, and history of cancer in the family. Then a complete physical examination, including general, breast, pelvic, and rectal examination, must be performed. 1 Prior to surgery a chest radiograph should be taken to screen for a pleural effusion and a CT scan of the abdomen and pelvis should be performed to delineate the extent of intra‐abdominal disease. However, in the absence of extra‐abdominopelvic disease, radiological scanning does not replace surgical staging with laparotomy. Tumor markers including CA125, and carcinoembryonic antigen (CEA) should be considered. 1  With a high CA125 level, the most common diagnosis would be epithelial ovarian, fallopian tube, or peritoneal cancer. A gastric or colonic primary with metastases to the ovaries may mimic ovarian cancer, and if CEA or CA19‐9 are elevated, this should be considered. A ratio of more than 25:1 (CA125 and CEA) favors an ovarian primary although it does not completely rule out a primary in the gastrointestinal tract. 51 A current mammogram should be considered as patients are frequently in the age group where breast cancer is prevalent. A colonoscopy is indicated when symptoms suggest possible colorectal cancer. 1 The following factors point to the presence of a malignancy, and are useful in the clinical assessment of masses: Age of the patient (young for germ cell, older for epithelial malignancies). Bilaterality. Tumor fixation clinically. Ascites. Ultrasonographically complex, especially if solid areas. CT finding of metastatic nodules. Elevated tumor markers. Age of the patient (young for germ cell, older for epithelial malignancies). Bilaterality. Tumor fixation clinically. Ascites. Ultrasonographically complex, especially if solid areas. CT finding of metastatic nodules. Elevated tumor markers.

To date, there are no documented effective screening methods that reduce the mortality of ovarian, fallopian tube, or peritoneal cancers. Studies using CA125, ultrasonography of the pelvis, and pelvic examination do not have an acceptable level of sensitivity and specificity, based on trials carried out in women in the general population 39 , 40 and those in the high‐risk population. 41 , 42  The US Preventive Services Task Force recommends against screening asymptomatic women for ovarian cancer with pelvic examination, pelvic ultrasound, or serum tumor marker measurements. 43  The low prevalence of disease and lack of high‐quality screening methods make it more likely to obtain false‐positive results leading to unnecessary interventions. A recent study of multimodal screening using CA125 based on a risk of ovarian cancer algorithm (ROCA) every 4 months and transvaginal ultrasound annually or earlier where indicated by the ROCA in women at high risk of ovarian cancer reported that screening was associated with a low rate of high‐volume disease at primary surgery and very high rates of no residual disease after surgery. 44 Given that the majority of women with advanced stage ovarian cancer, even with complete resection, will relapse after chemotherapy, this does not seem to be a good alternative to risk‐reducing surgery. The authors of the screening study concluded that risk‐reducing salpingectomy‐oophorectomy remains the treatment choice for women at high risk of ovarian/fallopian tube cancer. 44 Women at increased genetic risk should be encouraged to consider risk‐reducing bilateral salpingo‐oophorectomy, as this is the most effective way to reduce mortality in this population of women. 40 , 41 A bulletin from the American College of Obstetricians and Gynecologists (ACOG) has recommended that opportunistic (at the time of a clinically indicated hysterectomy) bilateral salpingectomy be considered in women not at genetic risk who wish to retain their ovaries as a way to reduce their risk of later developing high‐grade serous carcinomas. 45

A second‐look laparotomy (or laparoscopy) was previously performed in patients who have no clinical evidence of disease after completion of first‐line chemotherapy to determine response to treatment. Although of prognostic value, it has not been shown to influence survival, and is no longer recommended as part of the standard of care. 107 Level of Evidence C . Secondary cytoreduction is defined as an attempt at cytoreductive surgery at some stage following completion of first‐line chemotherapy. Retrospective studies suggest that patients benefit if all macroscopic disease can be removed, which usually means patients with a solitary recurrence. Patients with a disease‐free interval longer than 12–24 months and those with only 1–2 sites of disease appear to derive most benefit. 108 , 109  The role of secondary cytoreductive surgery is being evaluated in randomized clinical trials. The role of secondary debulking surgery has been addressed in the DESKTOP III trial and the results recently presented on behalf of the AGO. 110  This study included patients with a progression‐free survival of greater than 6 months after first‐line chemotherapy and who were considered to be good candidates for surgery based on a positive AGO Study Group score, defined as an ECOG performance status score of zero, ascites of 500 mL or less, and complete resection at initial surgery. Du Bois et al. 110 reported that the median progression‐free survival in 204 women who met these criteria and who were randomized to undergo surgery followed by chemotherapy was 18.4 months, compared with 14 months in 203 women who were randomized to receive only second‐line chemotherapy. Median overall survival showed an overall survival benefit of more than 12 months for patients undergoing complete secondary cytoreduction (60.7 months vs 46.2 months). Overall survival for patients who underwent surgery and were only incompletely cytoreduced was only 28 months, stressing the importance of complete cytoreduction. Results of the GOG 213 trial, however, showed no statistically significant difference in progression‐free survival of 18.9 months versus 16.2 months, and overall survival of 50.6 months versus 64.7 months (with vs without secondary cytoreduction). 111 In the view of these two trials, secondary cytoreduction can be considered a safe option for carefully selected patients. Level of Evidence B .

This group of ovarian tumors consists of a variety of histologically different subtypes that are all derived from the primitive germ cells of the embryonic gonad. Malignant germ cell tumors represent a relatively small proportion of all ovarian tumors. Prior to advances in chemotherapy, the prognosis for these aggressive tumors was poor. The use of platinum‐based chemotherapeutic regimes has made germ cell malignancies among the most highly curable cancers. 161 These are most common ovarian tumors in the second and third decades of life. They are frequently diagnosed by finding a palpable abdominal mass in a young woman who complains of abdominal pain. The following are the symptoms of germ cell tumors in order of frequency 161 : Acute abdominal pain. Chronic abdominal pain. Asymptomatic abdominal mass. Abnormal vaginal bleeding. Abdominal distention. Acute abdominal pain. Chronic abdominal pain. Asymptomatic abdominal mass. Abnormal vaginal bleeding. Abdominal distention. The classification of germ cell tumors of the ovary is important to determine prognosis and for treatment with chemotherapy. Germ cell tumors are classified as follows 2 , 161 : Dysgerminoma. Embryonal carcinoma Teratoma (immature, mature, mature with carcinoma [squamous cell, carcinoid, neuroectodermal, malignant struma, etc]). Extra‐embryonal differentiation (choriocarcinoma, endodermal sinus tumor [yolk sac tumor]). Dysgerminoma. Embryonal carcinoma Teratoma (immature, mature, mature with carcinoma [squamous cell, carcinoid, neuroectodermal, malignant struma, etc]). Extra‐embryonal differentiation (choriocarcinoma, endodermal sinus tumor [yolk sac tumor]). Ovarian germ cell tumors are staged similarly to epithelial carcinomas, although the staging system used for male germ cell tumors is probably more useful. The approach to treatment is based on the principles of management of metastatic germ cell tumors of the testis (i.e. low, intermediate, and poor risk). Dysgerminoma is the equivalent of seminoma in testicular cancer. 174 It is exquisitely sensitive to platinum‐based chemotherapy and is radiosensitive. The cure rate is high irrespective of the stage. The other histologic subtypes are equivalent to nonseminomatous testicular cancer. The aggressiveness of the disease is dependent on the type—the most aggressive being endodermal sinus and choriocarcinoma, but with combination chemotherapy, they are highly curable. 175 , 176 , 177 , 178 , 179 As chemotherapy can cure the majority of patients, even with advanced disease, conservative surgery is standard in all stages of all germ cell tumors. Conservative surgery means laparotomy with careful examination and biopsy of all suspicious areas, with limited cytoreduction, thereby avoiding major morbidity. The uterus and the contralateral ovary should be left intact. Wedge biopsy of a normal ovary is not recommended as it defeats the purpose of conservative therapy by potentially causing infertility. Patients with advanced disease may benefit from 3–4 cycles of neoadjuvant chemotherapy using BEP (bleomycin, etoposide, cisplatin [platinum]) regimen with preservation of fertility. 180  Patients who receive conservative surgery with the preservation of one ovary retain acceptable fertility rates despite adjuvant treatment with chemotherapy. There has been no report of higher adverse obstetric outcome or long‐term unfavorable sequelae in the offspring. Secondary surgery is of no proven benefit, except in those patients whose tumor was not completely resected at the initial operation and who had teratomatous elements in their primary tumor. Surgical resection of residual masses may be beneficial in such patients, as there may be mature teratomatous nodules that can continue to increase in size (growing teratoma syndrome), and more rarely can undergo malignant transformation over time to an incurable malignancy (e.g. squamous cell carcinoma). 181 Patients with Stage IA disease may be observed after surgery. A small proportion of patients may recur, but they can be treated successfully at the time of recurrence with a high rate of cure. Patients with disease beyond the ovary should receive adjuvant chemotherapy. Although radiation therapy is effective, it is no longer used in view of late effects and chemotherapy is highly effective. A follow‐up surveillance regime for patients with Stage IA dysgerminoma is outlined in Table  5 . This schedule is based on the experience managing seminomas in males and the reports by Dark et al. 182 and Patterson et al. 183  This pragmatic follow‐up schedule and has not been tested in randomized trials. Follow‐up regime for Stage I germ cell malignancies a Abbreviations: AFP, alpha‐fetoprotein; hCG, human chorionic gonadotropin; LDH, lactate dehydrogenase. Adapted from Patterson et al. 183 Dysgerminoma is extremely sensitive to chemotherapy and treatment with chemotherapy cures the majority of patients, even with advanced disease. 161 , 184  The recommended chemotherapy regimen is as follows: Etoposide (E) 100 mg/m 2 IV per day for 5 days every 3 weeks for 3 cycles. Cisplatin (P) 20 mg/m 2 IV per day for 5 days every 3 weeks for 3 cycles. Bleomycin (B) 30 IU IV/IM on days 1/8/15 for 12 weeks (optional) (Note: bleomycin is dosed in International Units). If bleomycin is omitted, then 4 cycles of EP are commonly used. (Note that various schedules of bleomycin have been used and the role of bleomycin in dysgerminomas is controversial). Etoposide (E) 100 mg/m 2 IV per day for 5 days every 3 weeks for 3 cycles. Cisplatin (P) 20 mg/m 2 IV per day for 5 days every 3 weeks for 3 cycles. Bleomycin (B) 30 IU IV/IM on days 1/8/15 for 12 weeks (optional) (Note: bleomycin is dosed in International Units). If bleomycin is omitted, then 4 cycles of EP are commonly used. (Note that various schedules of bleomycin have been used and the role of bleomycin in dysgerminomas is controversial). There is increased interest in de‐escalation of chemotherapy in dysgerminomas as they are so chemosensitive. It may be possible to omit bleomycin and substitute carboplatin for cisplatin due to the acute adverse effects and potential long‐term adverse effects associated with BEP, which include secondary malignancies, cardiovascular disease, hypertension, Raynaud's phenomenon, pulmonary toxicity, nephrotoxicity, neurotoxicity, deafness, decreased fertility, and psychosocial problems amongst others. GOG 116 is an old trial that investigated carboplatin 400 mg/m 2 and etoposide 120 mg/m 2 on days 1–3 every 4 weeks in 39 patients with Stage IB–III dysgerminoma. 185  No patients relapsed despite the very modest dose of carboplatin and 3 days of etoposide every 4 weeks for 3 cycles only; but the trial closed early after the results of two trials in males with nonseminomatous testicular cancer reported inferior outcomes with carboplatin compared with cisplatin. Shah et al. 186 reported the results of pooled data from six trials (three pediatric and three adult) on behalf of the Malignant Germ Cell Tumor International Consortium (MaGIC), which included 126 patients with advanced stage (Stage IC–IV) dysgerminomas who were treated with either carboplatin‐ or cisplatin‐based chemotherapy. Survival outcomes were equivalent with a 96% 5‐year survival in both groups with no differences seen according to age (25 years old). Seven patients relapsed including two patients treated with carboplatin‐based chemotherapy and five with BEP, and all were salvaged. When there is bulky residual disease it is common to give 3–4 courses of BEP or EP chemotherapy. 187 Level of Evidence B . The optimal follow‐up schedule has not been clinically investigated in ovarian germ cancers and the frequency of visits and investigations is controversial. Patients who have Stage I tumors and are offered surveillance need to be seen regularly and one option is to utilize the follow‐up regimen presented above. 182  Patients who have had chemotherapy have a lower risk of recurrence and the frequency of CT scans can be reduced, which is similar to the approach for testicular germ cell tumors. 183 Each follow‐up visit should involve taking a medical history, physical examination, and tumor marker determination. Although tumor markers are important, radiological imaging is also pertinent, especially for patients whose tumor markers were not raised at diagnosis. CT or MRI scans should be performed as clinically indicated. 182 Patients who have not received chemotherapy should be followed closely. Ninety percent of relapses in these patients occur within the first 2 years. At relapse, with few exceptions, these patients can be successfully treated. 182 Level of Evidence D . These tumors are highly curable with chemotherapy, even with advanced disease. Patients with Stage IA grade 1–2 immature teratoma have a very good prognosis and should be only observed after primary conservative surgery. Adjuvant chemotherapy does not appear to add any survival benefit in this subgroup of patients. Although adjuvant chemotherapy has been routinely recommended to all other patients with Stage I nondysgerminomatous ovarian germ cell tumors, this approach has been challenged and there may be a role for close surveillance and chemotherapy reserved for the subset who relapse as this is the standard of care in males with apparent Stage 1 testicular cancers. All other patients with nondysgerminomas, and higher‐stage and higher‐grade immature teratomas, should receive postoperative adjuvant chemotherapy. 161 The recommended chemotherapy regimen is etoposide 100 mg/m 2 per day for 5 days with cisplatin 20 mg/m 2 per day for 5 days, and bleomycin at 30 IU IM/IV on days 1, 8, and 15 for a total of 12 weeks of treatment. For patients with good prognosis disease, 3 cycles of BEP are recommended, while patients with intermediate/poor risk disease should receive 4 cycles of BEP. 161 Patients who relapse after BEP may still attain a durable remission and cure with second‐line chemotherapy regimens such as paclitaxel–ifosfamide–cisplatin (TIP). 177 High‐dose chemotherapy and autologous marrow rescue may be considered in selected patients. These patients should be managed in specialized units. After chemotherapy, patients with metastatic immature teratomas can sometimes have residual masses, which are composed entirely of mature elements. These masses can grow (growing teratoma syndrome), and should be resected after the completion of chemotherapy. 188 Level of Evidence B . All patients should have alpha‐fetoprotein (AFP) and human gonadotropin (beta hCG) to monitor response to treatment. All patients treated with chemotherapy should be followed up with medical history, physical examination, and appropriate tumor markers in the same way as dysgerminomas. CT or MRI scans should be performed as clinically indicated. 159 Relapses in patients usually occur within the first 2 years after diagnosis 161 , 177 Level of Evidence D .

There is no evidence to show that intensive clinical monitoring during follow‐up after completion of primary surgery and chemotherapy with early initiation of chemotherapy in asymptomatic women with recurrent disease improves overall survival or quality of life. In asymptomatic patients with CA125 progression and small volume disease or no radiological evidence of recurrence, it is appropriate to delay starting chemotherapy. However, there may be a subset of patients who are suitable for secondary debulking surgery at the time of recurrence. The objectives of follow‐up include: Early recognition and prompt management of treatment‐related complications, including provision of psychological support. Early detection of symptoms or signs of recurrent disease. Collection of data regarding the efficacy of any treatment and the complications associated with those treatments in patients treated in clinical trials. Promotion of healthy behavior, including screening for breast cancer in patients with early‐stage disease, and screening for cervical cancer in patients having conservative surgery. Early recognition and prompt management of treatment‐related complications, including provision of psychological support. Early detection of symptoms or signs of recurrent disease. Collection of data regarding the efficacy of any treatment and the complications associated with those treatments in patients treated in clinical trials. Promotion of healthy behavior, including screening for breast cancer in patients with early‐stage disease, and screening for cervical cancer in patients having conservative surgery. There are no evidence‐based guidelines regarding the appropriate follow‐up schedule. During the first year following treatment, patients are seen every 3 months with a gradual increase in intervals to every 4–6 months after 2 years and then annually after the fifth year. At each follow‐up, the patient should have her history retaken, including any change in family history of cancers and attention to any symptoms that could suggest recurrence; a physical and pelvic examination should be performed. This is an opportunity to refer appropriate patients for genetic testing if it was not done at diagnosis or during treatment. CA125 has traditionally been checked at regular intervals, but there has been debate regarding the clinical benefit of using CA125 progression alone as a trigger for initiating second‐line chemotherapy. A large MRC OV05‐EORTC 55955 study showed that treating asymptomatic patients with recurrent ovarian cancer with chemotherapy on the basis of CA125 progression alone did not improve survival and early treatment in asymptomatic patients had a negative impact on quality of life. 112  This study has generated considerable debate regarding the use of CA125 for follow‐up, but most agree that it is reasonable not to immediately initiate treatment unless there is a clear clinical indication to do so. The timing of treatment should be based on symptoms as well as clinical and radiological findings. Imaging tests such as ultrasonography of the pelvis, CT, MRI, and/or positron emission tomography (PET) scans should be performed only when the clinical findings or the tumor markers suggest possible recurrence. There appears to be no benefit to initiating chemotherapy in an asymptomatic patient with recurrent disease based only on rising CA125 levels in the absence of clinical symptoms or radiological evidence of recurrence. In asymptomatic patients with small volume disease and no radiological evidence of recurrence, close observation is a reasonable option, as well as entry into an appropriate clinical trial or possibly a trial of tamoxifen may be considered. A Cochrane database systematic review of tamoxifen in unselected women with recurrent ovarian cancer reported a 10% objective response and a 32% disease stabilization rate. 113  The patients treated were heterogeneous and included asymptomatic patients with rising CA125 levels, and symptomatic patients with chemotherapy‐resistant disease who had been heavily pretreated and had a poor performance status. GOG 198 compared tamoxifen and thalidomide in women with recurrent Stage III or IV epithelial ovarian, tubal, or peritoneal cancer who had completed first‐line chemotherapy, and who subsequently had Gynecologic Cancer InterGroup (GCIG) documented CA125 progression. The study reported that women who received thalidomide had a 31% increased risk of disease progression (HR 1.31), compared with those who were given tamoxifen. 114  The median progression‐free survival was 3.2 months in the thalidomide group versus 4.5 months in the tamoxifen group. This suggests that tamoxifen may have a role in selected patients with a rising CA125 level, and the relationship between estrogen receptor positivity and benefit of tamoxifen in this patient population is being evaluated in current studies. In the PARAGON trial the role of anastrozole in 54 asymptomatic patients with rising CA125 was investigated in a phase 2 design. 115  The primary endpoint was clinical benefit at 3 months and this was observed in 18 patients (34.6%; 95% CI, 23%–48%). The median duration of clinical benefit was 6.5 months (95% CI, 2.8–11.7). Most patients progressed within 6 months of starting anastrozole but 12 (22%) continued treatment for longer than 6 months. The role of hormonal therapy in this setting remains uncertain.

The majority of patients who present with advanced epithelial cancers of the ovary, fallopian tube, and peritoneum will relapse with a median time to recurrence of 16 months. Patients with recurrent ovarian cancer constitute a heterogeneous group with a variable prognosis, and a variable response to further treatment. The most widely used clinical surrogate for predicting response to subsequent chemotherapy and prognosis has been the progression‐free interval or the “platinum‐free interval,” which is defined as the time from cessation of primary platinum‐based chemotherapy to disease recurrence or progression. 116 , 117  This has been useful to define specific patient populations, but it has a number of limitations and depends on how patients are followed. In particular, it depends on how recurrence is detected and defined. Patients with a treatment‐free interval of less than 6 months are classified as platinum resistant and generally treated with nonplatinum‐based chemotherapy, while those with a treatment‐free interval of more than 6 months are considered to be platinum sensitive and commonly treated with platinum‐based chemotherapy. Patients who progress while on treatment or within 4 weeks of stopping chemotherapy are classified as platinum refractory. 116 , 117 There have been modifications to these definitions, and time to progression or recurrence rather than treatment‐free interval or platinum‐free interval has been used to define specific patient populations. There has been significant change in practice over the last 20 years and patients have been routinely followed with regular CA125 testing after completion of chemotherapy. For example, the “platinum‐resistant” subgroup may include asymptomatic patients with CA125 progression alone at 3 months post chemotherapy or radiological evidence of recurrence as well as those who are symptomatic with clinical recurrence. The Fourth Ovarian Cancer Consensus Conference reached agreement that distinct patient populations should be based on the interval from last platinum therapy and the time to progression. The progression‐free interval is defined from the last date of platinum dose until progressive disease is documented. 116 , 117 For patients whose disease is considered platinum sensitive, the ICON4 study showed advantage in terms of overall survival and progression‐free survival for a combination of carboplatin and paclitaxel versus single‐agent carboplatin. 118 Level of Evidence A . For patients with neurotoxicity, gemcitabine 119 or liposomal doxorubicin 120  may be substituted for paclitaxel. A large GCIG study (CALYPSO) compared carboplatin and liposomal doxorubicin (CD) with carboplatin and paclitaxel (CP) in 976 patients. 121  The CD arm had statistically superior progression‐free survival compared with the CP arm, with a median progression‐free survival of 11.3 months versus 9.4 months, respectively. There was no significant difference in the overall survival between the treatment groups. Median overall survival was 33 months versus 30.7 months for the CP and CD arms, respectively. The CD arm was better tolerated with less severe toxicities, and this combination is now widely used. Level of Evidence A . There is evidence that the addition of bevacizumab to the regimen of carboplatin and gemcitabine improves progression‐free survival compared with carboplatin and gemcitabine in platinum‐sensitive disease. In the OCEANS study, 122 484 patients with platinum‐sensitive disease were randomly assigned to carboplatin (AUC 4 on day 1) and gemcitabine 1000 mg/m 2 on days 1 and 8) with or without bevacizumab (15 mg/kg on day 1) every 21 days cycles. Bevacizumab could be given concurrently with chemotherapy for a maximum of 10 cycles followed by bevacizumab alone until progression of disease or toxicity. The addition of bevacizumab to carboplatin and gemcitabine resulted in an improvement in progression‐free survival (12 months vs 8 months; HR 0.48; 95% CI, 0.39–0.61); however, there was no difference in overall survival between the two arms. Treatment with bevacizumab was associated with higher rates of serious hypertension (17% vs <1%), proteinuria grade 3 or higher (9% vs 1%), and noncentral nervous system bleeding (6% vs 1%). 122  The OV21 trial randomized 682 patients with platinum‐sensitive recurrent ovarian cancer to 6 intravenous cycles of bevacizumab (15 mg/kg, day 1) plus carboplatin (AUC 4, day 1) plus gemcitabine (1000 mg/m 2 , days 1 and 8) every 3 weeks (standard group) or 6 cycles of bevacizumab (10 mg/kg, days 1 and 15) plus carboplatin (AUC 5, day 1) plus pegylated liposomal doxorubicin (30 mg/m 2 , day 1) every 4 weeks (experimental group), both followed by maintenance bevacizumab (15 mg/kg every 3 weeks in both groups) until disease progression or unacceptable toxicity. The median progression‐free survival was 13.3 months (95% CI, 11.7–14.2) in the experimental group versus 11.6 months (95% CI, 11.0–12.7) in the standard group (HR 0.81; 95% CI, 0.68–0.96; P  = 0.012). 123  The results of this trial support the experimental regimen in clinical practice. For patients with definite platinum‐resistant disease, enrollment on available clinical trials or treatment with nonplatinum chemotherapy should be considered. There are a number of chemotherapy options including liposomal doxorubicin, 124 topotecan, 124 etoposide, 125 , 126 and gemcitabine. 127 , 128  The reported response rates are low, about 10%, with a median time to progression of 3–4 months and a median survival of 9–12 months. There have been many trials carried out with new agents in patients with platinum‐resistant ovarian cancer, including epothilones, 129 trabectedin, 130 and pemetrexed, 131 among others, with no significant increase in response rates or progression‐free survival. More recently there have been encouraging reports of novel new agents or combinations including Wee1 (WEE1hu) inhibitor adavosertib combined with gemcitabine, 132 as well as mirvetuximab soravtansine in patients with high folate receptor alpha expression, 133 and these drugs are actively being investigated. There are many clinical trials in progress for patients with platinum‐resistant ovarian cancer and these are listed on ClinicalTrials.gov. No new cytotoxic agent has been approved to treat recurrent ovarian cancer for many years. There is a role for angiogenesis inhibitors in platinum‐resistant ovarian cancer. In the AURELIA trial, women with recurrent platinum‐resistant ovarian cancer were randomized to standard of care, i.e. weekly topotecan, weekly paclitaxel, or monthly liposomal doxorubicin versus these agents combined with bevacizumab (10 mg/kg every 2 weeks, or 15 mg/kg every 3 weeks). 134  Patients in the experimental arm had a longer progression‐free survival of 6.7 months versus 3.4 months and a higher overall response rate of 30.9% versus 12.6%. An exploratory subgroup analysis noted an increase in overall survival for weekly paclitaxel plus bevacizumab from 13.4 months to 22.4 months (with and without bevacizumab). 135  The findings in the AURELIA trial changed the standard of care. There has been much interest in exploring the role of immune checkpoint inhibitors in patients with recurrent ovarian cancer including those with platinum resistance. However, in general the results of these studies have been disappointing with low response rates reported. For example, KEYNOTE‐100 evaluated pembrolizumab, an anti‐PD‐1 antibody, in patients with recurrent ovarian cancer after multiple prior lines. 136  The overall response rate was 8%, with a combined positive score (CPS, quantifying the number of PD‐L1 positive cells) over 10 the objective response rate was 11%–18%. Similarly, the response rate with avelumab, an anti‐PD‐L1 antibody, was 10% in recurrent ovarian cancer. 137 However, there may be a role for combination regimens, which are being explored. For example, the phase 1/2 TOPACIO trial using niraparib and pembrolizumab in recurrent platinum‐resistant ovarian cancer showed a response rate of 18%. 138  The combination of the CTLA‐4 antibody ipilimumab with nivolumab, an anti‐PD‐1 antibody induction, followed by nivolumab maintenance had an objective response rate of 31.4% compared with 12.2% with nivolumab alone in a recently reported randomized phase 2 trial. 139 Although the median progression‐free survival was longer with combination, it was only 3.9 months versus 2 months, and the benefit questionable given the increased toxicity. The multicohort Leap‐005 trial recently reported preliminary data on another combination treatment using pembrolizumab and the multityrosine kinase inhibitor lenvatinib. In 31 patients with recurrent ovarian cancer the response rate was 29%. 140  There are still more trials in progress that are likely to provide results over the next few years. It will take time to define the role of immune checkpoint inhibitors in patients with recurrent ovarian cancer, but it seems likely that only a small subset of patients benefits and the challenge is to identify who these patients are. The optimal management of a patient with platinum‐resistant or refractory disease is complex and requires a careful assessment of the patient's performance status, symptoms, and extent of disease. Attention to symptom control and good palliative care is an essential component of management. With very few exceptions, recurrent disease is not curable and the aim of treatment is to maintain quality of life and palliate symptoms particularly in patients with platinum‐resistant ovarian cancer. 141  There are many potential treatment options, including chemotherapy, angiogenesis inhibitors, radiation therapy, or surgery in selected patients and inclusion in clinical trials. There is a subset of patients who may benefit from secondary surgical debulking. Olaparib is US Food and Drug Administration (FDA) approved for the treatment of patients with germline BRCA‐mutated recurrent ovarian cancer who have received three or more prior lines of chemotherapy. The FDA granted approval on the basis of the response rate in a single‐arm study of olaparib in patients with BRCA mutations and with a wide range of different cancers. The response rate was 34% in heavily pretreated BRCA‐positive patients with platinum‐resistant recurrent ovarian cancer and the median progression‐free survival was 7.9 months. 142 Rucaparib is also approved for treatment of BRCA‐mutation‐associated advanced ovarian cancer after completion of treatment with two or more chemotherapy regimens regardless of whether patients are platinum sensitive or resistant. Rucaparib's approval was based primarily on efficacy data from 206 patients with BRCA‐associated recurrent ovarian cancer who had prior treatment with two or more chemotherapy regimens and safety data from 377 patients with ovarian cancer treated with 600 mg rucaparib orally twice daily. 143 Investigator‐assessed objective response rate was 54% and the median duration of response was 9.2 months. 143

Malignant tumors of the ovaries occur at all ages with variation in histologic subtype by age. For example, in women younger than 20 years of age, germ cell tumors predominate, while borderline tumors typically occur in women in their 30s and 40s—10 or more years younger than in women with invasive epithelial ovarian cancers, which mostly occur after the age of 50 years. The lifetime risk of a woman in the USA developing ovarian cancer is approximately 1 in 70. Approximately 23% of gynecologic cancers are ovarian in origin, but 47% of all deaths from cancer of the female genital tract occur in women with ovarian cancer. Overall, epithelial ovarian cancer accounts for 4% of all new cancer diagnoses in women and 5% of all cancer‐related deaths. 1 , 2 , 26 The overall incidence of epithelial tumors varies from 9–17 per 100 000 and is highest in high‐income countries, with the exception of Japan. 27 However, this incidence rate increases proportionately with age. The largest number of patients with epithelial ovarian cancer is found in the 60–64 years age group. The median age is about a decade earlier in low‐income countries. Established risk factors for epithelial ovarian tumors include reproductive risk factors. Women who have never had children are twice as likely to develop this disease. First pregnancy at an early age, early menopause, and the use of oral contraceptives have been associated with lower risks of ovarian cancer. 28  The relationship of these variables to fallopian tube cancer is unclear. As noted above, it has been previously presumed that fallopian tube malignancies were rare; however, this has been challenged by evidence to show that many tumors that were classified as serous carcinomas of the ovary or peritoneal cancers appear to have their origin in the fallopian tube. 3 , 4 , 5 , 6 , 7  When the origin is uncertain, the convention of designating all serous cancers as originating in the ovary should no longer be used and the term “undesignated origin” may be applied at the discretion of the pathologist. 19 Hereditary factors are implicated in approximately 20% of ovarian, fallopian tube, and peritoneal cancers 29 , 30 , 31 , 32 , 33 : Most hereditary ovarian cancers are due to pathogenic mutations in either the BRCA1 or BRCA2  genes. At least 15% of women with high‐grade nonmucinous ovarian cancers have germline mutations in BRCA1 / 2 and, importantly, almost 40% of these women do not have a family history of breast/ovarian cancer. All women with high‐grade nonmucinous invasive ovarian cancers should be offered genetic testing even if they do not have a family history of breast/ovarian cancer. Inherited deleterious mutations in BRCA1 and BRCA2 are the major genetic risk factors. Women who carry germline mutations in BRCA1 and BRCA2  have a substantially increased risk of ovarian, tubal, and peritoneal cancer—about 20%–50% with BRCA1 and 10%–20% with BRCA2 . 30 , 31 , 32 , 33  Typically, these cancers occur at an earlier age than sporadic cancers, particularly in BRCA1  mutation carriers, with a median age of diagnosis in the mid‐40s. A number of other low‐ to moderate‐penetrance genes can also predispose to ovarian, fallopian tube, or peritoneal cancer. A study using next generation sequencing of constitutional DNA samples from 1915 women with ovarian cancer was carried out to identify germline mutations using a panel of 20 genes including BRCA1 and BRCA2 , DNA mismatch repair genes, double‐stranded DNA break repair genes such as CHEK2 and ATM , as well as the BRCA1 ‐associated complex or the BRCA2 /Fanconi Anemia pathway genes (including BRIP1 , BARD1 , PALB2 , RAD50 , RAD51C , and RAD51D , among others). About 80% of mutations were in BRCA1 or BRCA2 . About 3% of patients carried mutations in the Fanconi Anemia pathway genes, while only 0.4% had mutations in mismatch repair genes. 34 In an earlier similar study that included 360 patients, 24% carried germline loss‐of‐function mutations including 18% in BRCA1 or BRCA2 and 6% in BARD1 , BRIP1 , CHEK2 , MRE11A , MSH6 , NBN , PALB2 , RAD50 , RAD51C , or TP53 . 35 , 36 Inherited mutations in the mismatch repair genes associated with Lynch syndrome type II. Women carrying these mutations have an increased risk of a number of cancers including colon, endometrial, and ovarian cancer. Typically, the ovarian cancers that occur are endometrioid or clear cell histologically and are usually Stage I. 36 Most hereditary ovarian cancers are due to pathogenic mutations in either the BRCA1 or BRCA2  genes. At least 15% of women with high‐grade nonmucinous ovarian cancers have germline mutations in BRCA1 / 2 and, importantly, almost 40% of these women do not have a family history of breast/ovarian cancer. All women with high‐grade nonmucinous invasive ovarian cancers should be offered genetic testing even if they do not have a family history of breast/ovarian cancer. Inherited deleterious mutations in BRCA1 and BRCA2 are the major genetic risk factors. Women who carry germline mutations in BRCA1 and BRCA2  have a substantially increased risk of ovarian, tubal, and peritoneal cancer—about 20%–50% with BRCA1 and 10%–20% with BRCA2 . 30 , 31 , 32 , 33  Typically, these cancers occur at an earlier age than sporadic cancers, particularly in BRCA1  mutation carriers, with a median age of diagnosis in the mid‐40s. A number of other low‐ to moderate‐penetrance genes can also predispose to ovarian, fallopian tube, or peritoneal cancer. A study using next generation sequencing of constitutional DNA samples from 1915 women with ovarian cancer was carried out to identify germline mutations using a panel of 20 genes including BRCA1 and BRCA2 , DNA mismatch repair genes, double‐stranded DNA break repair genes such as CHEK2 and ATM , as well as the BRCA1 ‐associated complex or the BRCA2 /Fanconi Anemia pathway genes (including BRIP1 , BARD1 , PALB2 , RAD50 , RAD51C , and RAD51D , among others). About 80% of mutations were in BRCA1 or BRCA2 . About 3% of patients carried mutations in the Fanconi Anemia pathway genes, while only 0.4% had mutations in mismatch repair genes. 34 In an earlier similar study that included 360 patients, 24% carried germline loss‐of‐function mutations including 18% in BRCA1 or BRCA2 and 6% in BARD1 , BRIP1 , CHEK2 , MRE11A , MSH6 , NBN , PALB2 , RAD50 , RAD51C , or TP53 . 35 , 36 Inherited mutations in the mismatch repair genes associated with Lynch syndrome type II. Women carrying these mutations have an increased risk of a number of cancers including colon, endometrial, and ovarian cancer. Typically, the ovarian cancers that occur are endometrioid or clear cell histologically and are usually Stage I. 36 Women with a strong family history of epithelial ovarian, fallopian tube, or peritoneal cancers, particularly if there is a documented germline BRCA mutation, are advised to have a risk‐reducing bilateral salpingo‐oophorectomy after appropriate counseling and at the completion of childbearing. All women who are suspected of carrying a BRCA germline mutation, based on family history or young age of diagnosis and a high‐grade serous or high‐grade endometrioid cancer, should be offered genetic testing. BRCA mutations may also occur in women without a family history of breast/ovarian cancer, and genetic testing should be considered in patients from ethnic groups where there is a high incidence of founder mutations (e.g. Ashkenazi Jewish ancestry), and in women with high‐grade serous cancers under the age of 70 years. Australian guidelines advise that all women with invasive epithelial ovarian cancer apart from mucinous cancers diagnosed under the age of 70 should be offered BRCA mutation testing independent of family history and histologic subtype. 37 In contrast, the Society of Gynecologic Oncology (SGO) and National Comprehensive Cancer Network (NCCN) guidelines recommend that all women diagnosed with ovarian, fallopian tube, or peritoneal carcinoma, regardless of age or family history, should receive genetic counseling and be offered genetic testing. 38  Women whose family history suggests Lynch syndrome type II should undergo appropriate genetic counseling and testing.

In 2014, FIGO’s Committee for Gynecologic Oncology revised the staging to incorporate ovarian, fallopian tube, and peritoneal cancer in the same system. Changing the staging system required extensive international consultation. The primary site (i.e. ovary, fallopian tube, or peritoneum) is designated, where possible. When it is not possible to clearly delineate the primary site, these should be listed as “undesignated”. 1 , 2 It has been presumed that fallopian tube malignancies were rare. 2 However, histologic, molecular, and genetic evidence shows that as many as 80% of tumors that were classified as high‐grade serous carcinomas of the ovary or peritoneum may have originated in the fimbrial end of the fallopian tube. 3 , 4 , 5 , 6 , 7 , 8  Therefore, the incidence of fallopian tube cancers may have been substantially underestimated. These new data support the view that high‐grade serous ovarian, fallopian tube, and peritoneal cancers should be considered collectively, and that the convention of designating malignancies as having an ovarian origin should no longer be used, unless that is clearly the origination site. It has been suggested that extrauterine tumors of serous histology arising in the ovary, fallopian tube, or peritoneum might be described collectively as “Müllerian carcinomas” 1 , 2 or “pelvic serous carcinomas”. 9  The latter tumor designation is controversial because some peritoneal tumors might arise in extrapelvic peritoneum. Therefore, the simple term “serous carcinoma" is preferred, and most of these are high‐grade serous carcinomas (HGSC). Although there has been no formal staging for peritoneal cancers, the FIGO staging system is used with the understanding that it is not possible to have a Stage I peritoneal cancer. Ovarian epithelial tumors may arise within endometriosis or cortical inclusions of Müllerian epithelium, likely a form of endosalpingiosis. These include low‐grade endometrioid carcinomas, clear cell carcinomas, borderline and low‐grade serous carcinomas, and mucinous carcinomas. These tumors are thought to evolve slowly from lower‐grade precursor conditions (endometriotic cysts, cystadenomas, etc) and are classified as type I tumors. 5 Fallopian tube carcinomas arise in the distal fallopian tube and the majority of these are high‐grade serous carcinomas. These are thought to evolve rapidly from more obscure precursors and are designated as type II tumors. 5 , 6  This latter group encompasses high‐grade endometrioid carcinomas and carcinosarcomas. All of these high‐grade carcinomas are nearly always associated with mutations in the TP53  gene. 5 The lymphatic drainage of the ovaries and fallopian tubes is via the utero‐ovarian, infundibulopelvic, and round ligament pathways and an external iliac accessory route into the following regional lymph nodes: external iliac, common iliac, hypogastric, lateral sacral, para‐aortic lymph nodes and, occasionally, to the inguinal nodes. 1 , 10 , 11 , 12  The peritoneal surfaces can drain through the diaphragmatic lymphatics and hence to the major venous vessels above the diaphragm. The peritoneum, including the omentum and pelvic and abdominal viscera, is the most common site for dissemination of ovarian and fallopian tube cancers. This includes the diaphragmatic and liver surfaces. Pleural involvement is also seen. Other extraperitoneal or extrapleural sites are relatively uncommon, but can occur. 1 , 10 , 11 , 12 After systematic pathologic analysis has excluded a tubal or ovarian site of origin, malignancies that appear to arise primarily on the peritoneum have an identical spread pattern, and frequently may involve the ovaries and fallopian tubes secondarily. These “peritoneal” tumors are thought to arise in endosalpingiosis. Although CT scans can delineate the intra‐abdominal spread of disease to a certain extent, ovarian, fallopian tube, and peritoneal cancers should be staged surgically. Operative findings determine the precise histologic diagnosis, stage, and therefore the prognosis, of the patient. 1 , 9 , 10 , 12 , 13 , 14 In selected patients with advanced stage disease, it may be appropriate to initiate chemotherapy prior to surgical intervention, and in these cases there should be histologic or cytologic confirmation of the diagnosis prior to starting neoadjuvant chemotherapy (see 5.2.2. below). Chest radiograms may serve as a screen for pleural effusions. As distant metastases are infrequent, there is no requirement for other radiological evaluation unless symptomatic. Serum CA125 levels may be useful in determining response to chemotherapy, but they do not contribute to staging. Fallopian tube involvement can be divided into three categories. In the first, an obvious intraluminal and grossly apparent fallopian tube mass is seen with tubal intraepithelial carcinoma (carcinoma in situ) that is presumed to have arisen in the fallopian tube. These cases should be staged surgically with a histologic confirmation of disease. Tumor extension into the submucosa or muscularis and to and beyond the serosa can therefore be defined. These features, together with the laterality and the presence or absence of ascites, should all be taken into consideration. 1 , 3 , 6 , 7 In the second scenario, a widespread serous carcinoma is associated with a tubal intraepithelial carcinoma. A visible mass in the endosalpinx may not be seen but the histologic findings should be noted in the pathology report since they may indicate a fallopian tube primary. Tumors obliterating both fallopian tube and ovary may belong to this group but whether a presumptive assignment of a tubal origin can be made in such cases is controversial given that tubal intraepithelial carcinoma cannot be confirmed. In the third scenario—risk‐reducing salpingo‐oophorectomy—tubal intraepithelial carcinoma may be the only finding. It should be reported as originating in the fallopian tube and managed accordingly. The majority of early serous cancers detected are found in the fallopian tube, irrespective of genetic risk. 15 , 16 The updated, revised FIGO staging system combines the classification for ovarian, fallopian tube, and peritoneum cancer. It is based on findings made mainly through surgical exploration (as outlined above). Table  1 presents the 2014 FIGO staging classification for cancer of the ovary, fallopian tube, and peritoneum. The equivalents within the Union for International Cancer Control (UICC) TNM classification are presented in Table  2 . FIGO staging classification for cancer of the ovary, fallopian tube, and peritoneum Cancer of the ovary, fallopian tube and peritoneum: FIGO staging (2014) compared with TNM classification a 1. The primary site—that is, ovary, fallopian tube, or peritoneum—should be designated where possible. In some cases, it may not be possible to clearly delineate the primary site, and these should be listed as “undesignated”. 2. The histologic type should be recorded. 3. The staging includes a revision of the Stage III patients and allotment to Stage IIIA1 is based on spread to the retroperitoneal lymph nodes without intraperitoneal dissemination, because an analysis of these patients indicates that their survival is significantly better than those who have intraperitoneal dissemination. 4. Involvement of retroperitoneal lymph nodes must be proven cytologically or histologically. 5. Extension of tumor from omentum to spleen or liver (Stage IIIC) should be differentiated from isolated parenchymal splenic or liver metastases (Stage IVB). Source: Prat J. 17 In addition to these changes, several other modifications of the former staging system have been made to better prospectively capture the data. Stage IC is now divided into three categories: IC1 (surgical spill); IC2 (capsule ruptured before surgery or tumor on ovarian or fallopian tube surface); and IC3 (malignant cells in the ascites or peritoneal washings). Stage IIC has been eliminated. The updated staging includes a revision of the Stage IIIC based on spread to the retroperitoneal lymph nodes alone without intraperitoneal dissemination because an analysis of these patients indicates that their survival is significantly better than those who have intraperitoneal dissemination. 18  This category is now subdivided into IIIA1(i) (metastasis ≤10 mm in greatest dimension), and IIIA1(ii) (metastasis >10 mm in greatest dimension). Stage IIIA2 is now “microscopic extrapelvic peritoneal involvement with or without positive retroperitoneal lymph node” metastasis. The wording of Stage IIIB has been modified to reflect the lymph node status. Stage IVB now includes metastases to the inguinal lymph nodes. NX: Regional lymph nodes cannot be assessed. N0: No regional lymph node metastasis. N1: Regional lymph node metastasis. NX: Regional lymph nodes cannot be assessed. N0: No regional lymph node metastasis. N1: Regional lymph node metastasis. MX: Distant metastasis cannot be assessed. M0: No distant metastasis. M1: Distant metastasis (excluding peritoneal metastasis). MX: Distant metastasis cannot be assessed. M0: No distant metastasis. M1: Distant metastasis (excluding peritoneal metastasis). The majority of cases of ovarian cancer are of epithelial origin. FIGO endorses the WHO histologic typing of epithelial ovarian tumors. It is recommended that all ovarian epithelial tumors be subdivided according to the classification given below. 19 The histologic classification of ovarian, fallopian tube, and peritoneal neoplasia is as follows: Serous tumors. Mucinous tumors. Endometrioid tumors. Clear cell tumors. Brenner tumors. Undifferentiated carcinomas (this group of malignant tumors is of epithelial structure, but they are too poorly differentiated to be placed in any other group). Mixed epithelial tumors (these tumors are composed of two or more of the five major cell types of common epithelial tumors. The types are usually specified). Cases with high‐grade serous carcinoma in which the ovaries and fallopian tubes appear to be incidentally involved and not the primary origin can be labeled as peritoneal carcinoma or serous carcinoma of undesignated site, at the discretion of the pathologist. Serous tumors. Mucinous tumors. Endometrioid tumors. Clear cell tumors. Brenner tumors. Undifferentiated carcinomas (this group of malignant tumors is of epithelial structure, but they are too poorly differentiated to be placed in any other group). Mixed epithelial tumors (these tumors are composed of two or more of the five major cell types of common epithelial tumors. The types are usually specified). Cases with high‐grade serous carcinoma in which the ovaries and fallopian tubes appear to be incidentally involved and not the primary origin can be labeled as peritoneal carcinoma or serous carcinoma of undesignated site, at the discretion of the pathologist. Epithelial tumors of the ovary and fallopian tube are further subclassified by histologic grading, which can be correlated with prognosis. This grading system does not apply to nonepithelial tumors. 20  Two grading systems are applied. For nonserous carcinomas (most endometrioid and mucinous), grading is identical to that used in the uterus, based on architecture with a one‐step upgrade if there is prominent nuclear atypia, as follows: GX: Grade cannot be assessed. G1: Well differentiated. G2: Moderately differentiated. G3: Poorly differentiated. GX: Grade cannot be assessed. G1: Well differentiated. G2: Moderately differentiated. G3: Poorly differentiated. Serous carcinomas are the most common in both the ovary and tube. More than 90% of fallopian tube carcinomas are serous or high‐grade endometrioid adenocarcinoma. Other cell types have been reported, but are rare. 1 , 2 , 21 Serous carcinomas are graded in a two‐grade system befitting their biology. High‐grade serous carcinomas, including both classic appearing and those with SET features (solid, endometrioid‐like, and transitional) carry a high frequency of mutations in TP53 . 22 , 23 , 24 Low‐grade serous carcinomas are often associated with borderline or atypical proliferative serous tumors, often contain mutations in BRAF and KRAS and contain wild‐type TP53 . Most “moderately differentiated” serous carcinomas carry mutations in TP53 and should be combined with the high‐grade tumors. 20 , 23 , 24 , 25 Nonepithelial cancers, although uncommon, are extremely important. These include granulosa cell tumors, germ cell tumors, sarcomas, and lymphomas. They are discussed below as separate entities. Metastatic neoplasms to the ovary, such as tumors arising in the breast, lower reproductive tract sites (cervix or uterine carcinomas) and gastrointestinal tract (signet ring cell [Krukenberg] carcinomas, low grade appendiceal or pancreaticobiliary mucinous tumors and other neoplasms) are graded and staged in accordance with their respective sites of origin. 1 , 2

Outside the submitted work, JSB reports institutional research funding received from Tesaro and ImmunoGen, and participation on a Merck Data Safety Monitoring Committee (MK‐7339‐001 ENGOT‐ov43). Outside the submitted work, MF reports institutional research grants received from AstraZeneca, Novartis, and Beigene; consulting fees from AstraZeneca, GSK, MSD, Lilly, Novartis, and Takeda; honoraria for lectures from AstraZeneca, GSK, and ACT Genomics; and participation on an AGITG Data Safety Monitoring Board. All other authors report no conflicts of interest.