Low-Grade Serous Ovarian Carcinoma: a Retrospective Study from a Tertiary Cancer Institute

Abstract

The aim of this study was to evaluate the clinico-pathological behaviour and treatment patterns of low-grade serous carcinomas (LGSC) of ovary treated at a regional cancer centre. A retrospective analysis was done for the histopathology-proven cases of low-grade serous ovarian carcinoma, treated at a tertiary cancer institute between January, 2010, and September, 2019. There were 28 patients identified from the medical records with low-grade serous ovarian carcinoma. Median age of the patients was 43 years [22–79 years]. Average BMI was 22.3 ± 4.0 kg/m² [range 15.2–31.2]. Twenty-one (75%) were parous and 7 (25%) were non-parous women. Median CA125 level was 188 IU/ml [range 6–14,187 IU/ml]. Ten (35.7%) patients had primary surgery elsewhere and 8 (80%) out of these patients had to undergo repeat staging. Fertility sparing surgery (FSS) was offered to 4 (14.3%) patients. Five (17.8%) patients had received neoadjuvant chemotherapy for advanced disease and poor performance status. Almost 82.2% (23) of the patients had no macroscopic residual disease at the primary surgery. According to International Federation of Obstetrics and Gynaecologists (FIGO) stage for ovarian carcinoma, there were 7 (25%), 6 (21.4%), 13 (46.4%), and 2 (7.1%) patients in the stages I, II, III, and IV respectively. Post-operative adjuvant chemotherapy was offered to 7 (25%), hormonal therapy (anastrozole/tamoxifen) to 7 (25%), and rest of 14 (50%) patients were under surveillance. Median follow-up time for the study group was 36 months. Overall survival (OS) and disease-free survival (DFS) at 2 years was 96.4% and 89.1%, respectively. Low-grade serous carcinomas of ovary differ biologically from high-grade serous ovarian carcinoma. Surgery is the cornerstone of the treatment. Further research is needed to understand the behaviour of these tumours for effective treatment strategies in future.

Background

Epithelial ovarian carcinoma consists of 75–90% of the total ovarian cancers. Based on the pathology, immunohistochemistry, and molecular genetic analysis, five main types have been identified. These include high-grade serous carcinomas (70%), endometrioid carcinomas (10%), clear-cell carcinomas (10%), mucinous carcinomas (3%), and low-grade serous carcinoma [LGSC] (< 5%). [1] LGSC are rare. These tumours usually follow a relatively indolent course. Frequently, a non-invasive serous borderline component (with or without micropapillary pattern) is most likely to represent the progression of serous borderline tumours beyond microinvasion (≈ 10%). [2] Commonly encountered mutations in LGSCs are BRAF or KRAS mutations (38% and 19%, respectively). [3] According to the recent GLOBOCAN 2020, in India 45,701 (6.7%) new cases of ovarian cancers were being diagnosed every year. [4] Indian data on low-grade serous ovarian carcinomas is fragmented. The present study aims to audit the experience in the management of the patients diagnosed with low-grade serous carcinomas of ovary at a regional cancer institute.

Methodology

A single-institution retrospective observational study was conducted at a tertiary cancer centre. The study was approved by the departmental scientific board. Between January, 2010, and September, 2019, medical records with the pathological diagnosis of low-grade ovarian carcinoma were reviewed. Patient characteristics, clinical features, and surgical and pathological details were noted along with the patterns of adjuvant treatments.

Statistical Analysis

Analysis was performed using SPSS for Windows (version 26; SPSS Inc., Chicago, IL, USA). Normality of the quantitative data was checked using Kolmogorov–Smirnov test. Discrete categorical data presented as n (%). Qualitative or categorical variables represented as frequencies and proportions. Descriptive statistics like mean, median, standard deviation, and percentage of various parameters/variables were calculated. Overall survival was calculated from the date of initial diagnosis or primary surgery until death or censored at the point of last contact on follow-up. Disease-free survival was calculated from the date of primary surgery or NACT to the date of recurrence. Patients who were alive without recurrence at the time of analysis were censored. Survival data was plotted using Kaplan–Meier survival plots.

Results

Twenty-eight patients with the diagnosis of LGSC were identified. Majority (75%) were parous women (21/28). Unilateral ovarian involvement was seen in 17 (60.7%) patients, whereas bilateral ovaries were involved in 11 (39.3%) patients. Patient demographics, clinical presentation, and baseline characteristics are shown in Table 1.

Table 1.

Patient demographics and clinical features

	Median	Mean ± std deviation	Range
1. Age at diagnosis	43	43.4 ± 14.1	22–79
2. BMI	22.7	22.7 ± 4.0	15.2–31.2
3. Pre-treatment Ca 125 level	188	1050 ± 3114	6–14,187
4. Serum albumin level at first visit	4.2	4.1 ± 0.4	3.5–5.3
Clinical characteristics		No. of women,—n (%)
1. Parity    a. Non-parous    b. Parous		7 (25%) 21 (75%)
2. Pre-operative ascites		5 (17.8%)
3.Pre-operative pleural effusion		1 (3.6%)
4.Performance status (ECOG)    a. ECOG 0–1    b. ECOG 2–3		25 (89.3%) 3 (10.7%)
5. Elevated Ca 125 levels		23 (82.1%)

BMI, body mass index; ECOG, Eastern Cooperative Oncology Group

Mode of primary surgery was laparoscopy for 4 (14.3%) women, who were operated outside the institute initially and then referred. Unilateral cystectomy as a part of fertility sparing surgery without proper staging was done in three women. Two of these women were restaged at our institute using laparoscopic surgery and one through laparotomy. The fourth patient was 52 years old and also had initial laparoscopic hysterectomy with bilateral salpingo-oophorectomy along with appendicectomy and cholecystectomy. She was referred with histopathology, revealing bilateral low-grade ovarian tumour. She had undergone restaging with omentectomy and nodal dissection, and there was no macroscopic residual disease after surgery. However, final histopathology revealed omental metastasis and isolated metastasis in pelvic lymph nodes. She was started on hormonal therapy. Table 2 depicts the surgical details of the patients in the present study.

Table 2.

Surgical details

Details	N (%)
1. Mode of initial surgery    a. Open surgery (primary and interval)    b. Laparoscopic surgery	24 (85.7%) 4 (14.3%)
2. Outside operated    a. Repeat surgery (those operated elsewhere)      i. For complete staging      ii. For recurrent disease	10 (35.7%) 8/10 (80%) 5/10 (50%) 3/10 (30%)
3. FSS (fertility sparing surgery)	4 (14.3%)
4. NACT (2–3 cycles paclitaxel + carboplatin) followed by interval debulking surgery (IDS)	5 (17.8%)
5. Residual disease (based on [CC] complete cytoreductive scores)    a. CC0    b. CC1    c. CC2    d. CC3	23 (82.2%) 1 (3.6%) 3 (10.7%) 1 (3.6%)
6. FIGO staging (surgico-pathological)    a. Stage I    b. Stage II    c. Stage III    d. Stage IV	7 (25%) 6 (21.4%) 13 (46.5%) 2 (7.1%)

There were 7 (25%) nulliparous women in the study group. Four women were desirous of future fertility as they were in the reproductive age group and were offered fertility sparing surgery. Three of these women (one unmarried and two married nullipara) had initial laparoscopic cystectomy elsewhere without adequate staging. Restaging was performed with unilateral salpingo-oophorectomy and omentectomy in all three of them and small peritoneal nodule excision was done in two of them. These three women were advised family completion and kept on regular surveillance with no evidence of disease at the last follow-up. The fourth patient presented to the institute with complex cystic lesion along with the evidence of torsion. She underwent emergency laparotomy with unilateral salpingo-oophorectomy. Intra-operatively tumour rupture with spillage of necrotic contents was observed. The frozen section showed borderline serous tumour, whereas post-op histopathology revealed LGSC with capsular breach (stage 1c3). Comprehensive staging with post op adjuvant chemotherapy, 6 cycles of carboplatin were given. This patient later conceived and delivered a healthy baby post treatment and is on regular follow-up without any evident disease since last 8 years.

Almost 95.6% (22/23) women who had primary surgery (n = 23) underwent complete cytoreduction (CC0). Five (17.8%) women in total had received neo-adjuvant chemotherapy (NACT) in view of advanced disease and for poor performance status as per the ECOG (Eastern Cooperative Oncology Group) score. The chemotherapy given was a combination of paclitaxel and carboplatin. Three women with poor performance status who received NACT were elderly with age above 55 years [56, 69, and 79 years old, respectively]. There was one mortality in the NACT group, who was 79 years old, had a pelvic recurrence within 6 months of completion of the treatment, and succumbed prior to starting her second line of chemotherapy. There were two patients who were young [24 and 26 years, respectively] but had advanced disease at presentation and thus were given NACT. The details describing the individual presentations, indication, and clinical course of all the patients who underwent NACT are depicted in Table 3.

Table 3.

Details of patients undergoing NACT (N = 5)

Sr. no	Age (year)	Clinical presentation	NACT indication	Chemo given (NACT)	Surgical details (IDS)	CC score	Adjuvant Rx	DFI in months	Rx on recurrence	Status on last follow-up
1	24	Abdominal distension, CT scan showing multiple peritoneal deposits with small bowel and mesenteric deposits (largest 3 cm)	Advanced disease with possibility of multiple bowel resections, (FNAC low grade carcinoma—IHC primary ovarian)	3TC	TAH + BSO + TO + RSRA cyto- reduction (sub-optimal) (I/O deposits all over liver surface, subdiaphragm, small bowel loops, and its mesentery)	CC3	Chemo (3TC)	18 months, Referred at recurrence- Had multiple liver subcapsular mets, ileal serosal, and mesenteric and peritoneal deposits, proliferative growth occluding 90% of lumen at sigmoid colon Colonoscopy biopsy—recurrent serous carcinoma from on IHC-primary ovarian	20CRS End colostomy + radical cystectomy + ileal conduit	Lost follow-up after 20CRS (OS—26 months)
2	26	Abdominal and pelvic mass, fixed to lower abdominal wall	Advanced stage Fixed mass FNAC from mass positive for malignancy	3TC	Interval cytoreduction (TAH + BSO + TO + bladder and POD peritonectomy) Post-op had B/L HUN, stenting done	CC2	Chemo 3TC	69 months, had pelvic mass, mild ascites, multiple peritoneal and abdominal wall deposits, grade 3 bilateral renal parenchymal disease, had to undergo bilateral PCN	2nd line chemotherapy carboplatin 6 cycles	AWD (OS—87 months)
3	56	Gross ascites Cytology negative 3 times, FNAC from pelvic mass + ve	Poor PS ECOG 2–3	2TC	TAH + BSO + TO + pseudo Capsule excision	CC0	Chemo (4TC)	32 months	N/A	NED (OS—32 months)
4	69	Post hysterectomy, Abdominal distension and pain Imaging—pleural effusion, ascites, omental caking	Poor PS ECOG 2–3 Pleural effusion positive for malignancy	3TC	BSO + TO + Tumour debulking	CC1	Chemo 3TC  + Bev maintenance i/v/o persistent ascites	37 months	N/A	NED (OS—37 months)
5	79	Gross ascites (cytology positive) with omental caking on imaging	Poor PS ECOG 2–3	3 TC	Total hysterectomy + BSO + infra colic omentectomy + excision of necrotic POD deposit	CC2	3TC	6 months	Had pelvic recurrence Was planned for 2nd line chemo Succumbed before treatment	Dead with disease (OS—10 months)

NACT, neoadjuvant chemotherapy; DFI, disease−free survival; CC score, complete cytoreduction score; OS, overall survival; DFI, disease−free interval; IDS, interval cytoreductive surgery; Rx, treatment; TC, paclitaxel and carboplatin combination regime; Bev, bevacizumab; 2°CRS, secondary cytoreductive surgery; AWD, alive with disease; NED, no evidence of disease, PS, performance status; ECOG, Eastern Co−operative Oncology Group; FNAC, fine needle aspiration cytology; TAH, total abdominal hysterectomy; BSO, bilateral salpingo−oophorectomy; TO, total omentectomy; RSRA, recto−sigmoid resection anastomosis; POD, pouch of Douglas; IHC, immuno−histochemistry; HUN, hydroureteronephrosis; PCN, percutaneous nephrostomy

Treatment patterns were observed to be inconsistent with respect to adjuvant or maintenance therapy in the post-operative period. Figure 1 shows that there was an observed predilection towards giving adjuvant chemotherapy for advanced stages and hormonal treatment for early stages in post-operative period. Bevacizumab was given to one patient who initially had received NACT (in view of pleural effusion with positive cytology) followed by interval debulking surgery. She had a suboptimal cytoreduction so she was given 3 more cycles of adjuvant chemotherapy. She was planned for bevacizumab in view of ascites which responded well with the same and was under remission at the end of the study period.

Fig. 1.

Treatment trends for adjuvant therapy

Median follow-up time was found to be 36 months [mean ± SD = 47 ± 33.8 months, range 10–100 months]. A total of 7 (25%) patients had developed recurrence during the entire study period. Among them, 2 (7.1%) patients died of disease and one patient with advanced disease was lost to follow-up. Table 4 highlights the individual characteristic findings in each of the patient who had developed recurrence of the disease. It is noteworthy that the majority of patients who had disease recurrence had advance stages at presentation. Those with early stages at presentation had significantly long disease-free intervals (DFI), emphasising further the indolent nature of these tumours. Overall survival (OS) and disease-free survival (PFS) at 2 years were 96.4% and 89.1%, respectively. Five years OS and DFS in the study group were observed to be 86.7% and 84.5%, respectively. Figure 2 depicts the Kaplan–Meier survival plot for overall survival (OS) and disease-free survival (DFS) among the patients with LGSC in this study.

Table 4.

Details of patients with recurrent disease (N = 7)

Sr. no	Age	Primary Rx	Surgico-pathological stage (initial Sx)	CC score	DFI/PFI (months)	Site of recurrence	Rx of recurrence	OS in months	Status at the end of study period
1	32 years	Sx	3c	CC 0	24	Vault/peritoneal	Chemotherapy (3 lines) +  pelvic RT + hormonal Rx	80	AWD
2	33 years	Sx + CT (6 T + C)	3c	CC 2	14	Peritoneal Had subacute intestinal obstruction, pleural effusion, hydropneumothorax	2oCRS Chemotherapy (3 lines) +  Hormonal Rx	40	Dead with disease
3	79 years	NACT (3 T + C) f/b Sx + Adj CT	3c	CC 2	6	POD	Chemotherapy was planned but succumbed prior to Rx	10	Dead with disease
4	53 years	Sx	1a*	CC 0	156	Other side ovary	Sx	173	NED
5	53 years	Sx	2b*	CC 0	84	Inguinal nodes	Sx (IFLND) + chemotherapy	99	NED
6	26 years	NACT (3 T + C) + Sx + adjuvant CT	3c	CC 2	69	Peritoneal/colon	Chemotherapy	87	AWD
7	24 years	NACT (3 carboplatin) + Sx + chemo	4b	CC 3	18	Pelvic mass	Sx (2°CRS) + chemotherapy	26	±Lost to follow-up

DFI, disease−free survival; PFI, progression−free interval; CC score, complete cytoreduction score; OS, overall survival; Sx, surgery; CT, chemotherapy; NACT, neoadjuvant chemotherapy; POD, pouch of Douglas; Rx, treatment; 2°CRS, secondary cytoreductive surgery; CT, chemotherapy; T + C, carboplatin + paclitaxel; RT, radiotherapy; AWD, alive with disease; NED, no evidence of disease; IFLND, inguino−femoral lymphadenectomy

*Early stages at presentation with long DFI, ± lost to follow−up—had advanced disease and alive with disease on last follow−up visit

Fig. 2.

Kaplan–Meier survival plot showing overall survival and disease-free survival of the patients with LGSC from the study

Discussion

The present study analysed the data of patients diagnosed with LGSC of ovary compiled from a tertiary care cancer centre in India. Low-grade serous carcinoma (LGSC) of ovary constitutes a small group of epithelial ovarian cancers (EOC), almost < 5% all EOC. [1] Commonly diagnosed in the advanced stage, their treatment is also on the similar lines as that of their high-grade variant. The LGSC arises de novo or from the serous borderline tumours. The present WHO classification considers serous borderline ovarian tumour with invasive implants as low-grade serous cancers. [5] This has led to a whole set of the cohort being reclassified into LGSC. A retrospective series from the MD Anderson Cancer Centre found that about 60% of LGSC were observed in the context of a serous borderline tumour at diagnosis. [6] Pathological evaluation in the present study showed that there was co-existing borderline tumour either in the same or the opposite ovary in 6 (21.4%) patients. Specimen from 2 patients revealed tiny foci of high-grade changes. Malignant transformation from low-grade to high-grade serous ovarian cancer is described in the literature, although it is difficult to explain on the molecular basis. [7]

The present study demonstrated younger age at presentation (median age, 43 years) for LGSC of ovary, which has been substantiated in previous cohort studies as well. [8] Because of the younger age at diagnosis, fertility preservation is of utmost concern. Fertility sparing surgery can be safely performed for FIGO stages IA and IC1 of LGSC. [9, 10] Fertility sparing option must therefore be discussed with the eligible patient and her family.

A major number (n = 23, 82.2%) of patients in our cohort had cytoreductive score (CRS) of CC0. Among these was a patient who had received 2 cycles of NACT for poor performance status and was operated thereafter. Studies have shown that LGSC of ovary are not very responsive to chemotherapy as compared to high-grade serous cancers. [11, 12] The Arbeitsgemeinschaft Gynaekologische Onkologie Studiengruppe Ovarialkarzinom (AGO-OVAR) group investigated the response rate to chemotherapy in LGSC (n = 39) and HGSC (n = 80) having sub-optimal debulking (RD > 1.0 cm). This was an exploratory case control study of 4 randomised phase III trials of first-line chemotherapy from the AGO-OVAR database. Observed response rates to chemotherapy were 23.1% and 90.1% in LGSC and HGSC. They also noted that when complete cytoreduction was achieved, both the groups had significantly better PFS and OS in comparison to those who had residual disease (p < 0.001). [12] In their retrospective study of patients with advanced LGSC, Bogani et al. analysed 68 patients who underwent primary cytoreductive surgery. [13] They observed that non-optimal cytoreduction (hazard ratio [HR] = 2.79; 95% confidence interval [CI] = 1.16–6.70; p = 0.021) and International Federation of Obstetrics and Gynaecologists (FIGO) stage IV (HR = 3.15; 95% CI = 1.29–7.66; p = 0.011) were associated with worse disease-free survival (DFS) and increased risk of recurrence on multivariate analysis. They further concluded that interval debulking surgery with non-optimal cytoreductive surgery could be detrimental. In the present study, there were a total of 7 cases, who had a recurrence of disease during the study period. The patients with early stage at initial presentation had a significantly longer disease-free interval (DFI) compared to those presented with advanced stages at initial diagnosis. There were only two mortalities observed in the present study, that too in the recurrent cases. One patient with advanced disease with recurrence was lost to follow-up after secondary cytoreduction. This patient was initially treated with NACT elsewhere in view of advanced peritoneal and small bowel deposits and multiple adhesions. Interval surgery was suboptimal (CC3) as she had extensive deposits over liver surface, sub-diaphragm, small bowel loops, and its mesentery, also involving sigmoid colon (stage IV b). She underwent rectosigmoid resection and anastomosis followed by 3 more cycles of adjuvant chemotherapy. She had a stable disease until she progressed after 18 months.

In the present study, hormone receptor status was available only in very few patients owing to the cost factor. This came as a significant limitation. LGSC have significantly higher expression of estrogenic (ER) and progesterone (PR) receptors. [14] Gehrsenson et al., in their analysis of 203 eligible patients with LGSC, found that women with FIGO stages II to IV who were given hormonal maintenance therapy had a longer PFS [median PFS 64.9 months vs. 26.4 months, p < 0.001] compared with the women who were considered for observation and follow-up. [15] They also had a lower risk of progression among the subgroup with no disease or persistent disease (hazard ratio, 0.44; 95%CI, 0.31 to 0.64; P, 0.001). However, they did not observe any statistically significant benefit in the overall survival among these groups [102.7 vs. 115.7 months, respectively]. Present NCCN (National Comprehensive Cancer Network) guidelines recommend post-operative observation for stage IA and IB low-grade serous ovarian cancers, whereas for stage IC to II disease, post-operative observation or adjuvant chemotherapy (carboplatin + paclitaxel/docetaxel based) or hormonal therapy may be considered. [16] However, for FIGO stage III and IV disease, first-line chemotherapy as recommended for epithelial ovarian cancer followed by maintenance hormonal treatment/observation or a primary hormonal therapy (including anastrozole, letrozole, leuprolide, or tamoxifen) can be considered post-operative options. The present study could not provide any conclusion regarding the optimal hormonal maintenance therapy to be considered. A total of 10 patients were put on hormonal maintenance therapy, most of them received aromatase inhibitor (anastrozole) and 2 patients were on tamoxifen therapy. It was observed though that there was increasing tendency to start hormonal maintenance therapy recently, and also more for the early stages (Fig. 1). Siemon et al. recently examined the treatment differences between practitioners in LGSC. [17] The practice patterns they observed had a large degree of variation between gynaecologic oncologists across the USA. Associated factors for these preferences included patient volume, training region, practice settings, current practice, and also the years of experience. They emphasized the need for further efforts to improve knowledge of this disease, to identify optimal treatment modalities, and to provide guidelines for management.

Study Limitations

Smaller sample size of the cohort. Hormone receptor analysis was not available for majority of the patients in the present study group.

Conclusion

Low-grade serous carcinomas (LGSC) of ovary occur at an early age compared to their high-grade counterpart. Surgery forms an important cornerstone of treatment owing to their poor response to chemotherapy. They have an indolent course and prolonged survival despite the advanced stage at presentation. Further research is needed to understand the pathogenesis and to identify optimal treatment modality. Multi-centric trials or a national registry may help gather more information as these are rare tumours. Hormone receptor status evaluation must be done in every case to consider adjuvant hormonal maintenance therapy, owing to its proven efficacy in their treatment.

Author Contribution

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Dr. Rajasekhar and Dr. Sravanthi. The first draft of the manuscript was written by Dr. Sravanthi and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Declarations

Ethics Approval

Institutional ethical committee approval taken.

Conflict of Interest

The authors declare no competing interests.