Clinical case

A 36-year-old, G1P1, previously healthy woman presents with left sided pelvic pain. She presents to her PCP who performs a CBC and comprehensive panel that are unrevealing, abdominal exam shows bloating without point tenderness. She has a history of normal menses. She has no family history of cancer, and her both parents are alive and healthy. She is referred to her gynecologist who performs a transvaginal pelvic ultrasound which reveals bilateral multilocular ovarian cysts with mural nodularity, papillary projections and solid components. She is sent for a contrast CT scan of abdomen and pelvis which shows left ovarian mass consistent with an ovarian neoplasm, mild abdominopelvic peritoneal disease and small slightly complex right ovarian cyst ( Fig. 6.1 ). Dedicated chest CT shows clear lungs with no signs of metastatic disease to the chest. Blood tests show CA-125 of 125 U/mL.

Fig. 6.1, CT of abdomen and pelvis showing left ovarian mass in a 36 year old woman with newly diagnosed stage IIIc low-grade serous ovarian cancer.

She is seen in consultation by a gynecologic oncologist who recommends surgical staging. The patient is seen by reproductive endocrinology prior to surgery. She does not plan to have further children and decides to not attempt harvesting of oocytes prior to surgery. At time of surgery an open laparotomy is performed with TAH/BSO, omentectomy, washings, and resection of all gross disease. She is found to have involvement of the ovaries, left fallopian tube, uterine serosa, omentum, and abdominal and pelvic peritoneal disease, consistent with stage IIIc ovarian cancer. Pathology examination shows serous adenocarcinoma, arising in serous borderline disease with micropapillary features. The invasive component is p53 wild-type, ER +, PR-, WT-1, and PAX-8 positive, consistent with low-grade serous ovarian cancer.

The patient undergoes germline and somatic testing and is found to have no BRCA mutation but a somatic KRAS mutation. She is treated postoperatively with 6 cycles of carboplatin and paclitaxel chemotherapy followed by letrozole maintenance therapy. She is routinely followed with blood work and exam every 3 months and CT scan of chest, abdomen, and pelvis every 6 months. Three years into maintenance therapy she is found to have recurrent disease with an isolated peritoneal implant. She undergoes secondary debulking surgery with her gynecologic oncologist. Postoperatively she is treated with 6 cycles of carboplatin and gemcitabine, which she tolerates well, but at end of treatment repeat CT imaging shows recurrence of disease with multiple peritoneal nodules and supradiaphragmatic nodes. Biopsy confirms recurrent low-grade serous ovarian cancer. At this time, what systemic therapies would you consider and how may molecular testing help in your decision making?

Epidemiology

Incidence/mortality

Low-grade serous ovarian cancer is a rare form of epithelial ovarian, fallopian tube or primary peritoneal cancer, accounting for approximately 5%–7% of cases of epithelial ovarian cancer. Like the most common form of ovarian cancer, high-grade serous ovarian cancer (which accounts for ~ 70% of cases), low-grade serous ovarian cancer is typically diagnosed at an advanced stage. Patients with low-grade serous ovarian cancer tend to have a more protracted disease course with a median survival of approximately 10 years; however, patients have limited responses to systemic therapy with most ultimately succumbing to their cancer. A retrospective study by Bodurka et al. reviewed the data from 290 patients with stage III serous ovarian cancer treated with surgery and chemotherapy on GOG 158; blinded pathology review was performed by a panel of 6 gynecologic pathologists to regrade the tumors using the 2-tier (low-grade vs. high-grade) histologic classification system. Patients with low-grade serous ovarian cancer had a significantly longer PFS (45 vs 19.8 months respectively; P = 0.01) and women with high-grade serous ovarian cancer demonstrated a significantly higher risk of death vs those with low-grade serous (HR, 2.43; 95% CI, 1.17–5.04; P = 0.02). Patients with low-grade serous ovarian cancer also have a younger median age of diagnosis than those with high-grade serous ovarian cancer, with patients commonly diagnosed in their 20s or 30s ( Table 6.1 ).

Table 6.1
Key characteristics of low-grade vs high-grade serous ovarian cancer.
References: Grisham, RN. Oncology . 2016;7:650–2. J Natl Cancer Inst . 2014;106(4):1–8; Bodurka et al. Cancer . 2012;118:3087–94; Schmeler et al. Gynecol Oncol . 2008;108:510–4; Grabowski et al. Gynecol Oncol . 2016;140(3):457–462. Gershenson et al. Gynecol Oncol . 2009;114(1):48–52. Pujade-Lauraine et al. J Clin Oncol . 2014;32(13):1302–8. Aghajanian et al. J Clin Oncol 2012;30(17):2039–45.
Clinical/molecular features Low-grade serous High-grade serous
Median age at diagnosis 40–50 years 50–60 years
Molecular genetics Mutant: RAS, BRAF
Wild type: p53
Mutant: p53, BRCA
Wild type: BRAF, RAS
Response rate to neoadjuvant chemotherapy 4%–23% 80%–90%
Response rates to chemotherapy in the recurrent setting 2%–25% 12%–57%
Median PFS/OS for patients with Stage III optimally debulked disease treated on GOG 158 (blinded pathology review) n = 21
PFS: 45 months
OS: 126.2 months
n = 220
PFS: 19.8 months
OS: 53.8 months

Etiology/risk factors

Unlike high-grade serous ovarian cancer, the incidence of BRCA mutations in patients with low-grade serous ovarian cancer approaches that of the general population. There is not a known hereditary or environmental predisposition for development of this disease. The primary risk factor for development of invasive low-grade serous ovarian cancer is a personal history of its precursor lesion, serous borderline disease. Benign ovarian cystadenoma can progress in a step wise fashion to serous borderline tumor (SBT), SBT with micropapillary features, and ultimately invasive low-grade serous ovarian cancer ( Fig. 6.2 ). A comprehensive population study performed by Vang et al. involved the entire female population of Denmark and included all patients diagnosed with a SBT between 1978 and 2002 or with low-grade serous ovarian cancer between 1997 and 2002. In total 1042 cases of serous borderline disease were included; subsequent development of carcinoma occurred in 4% of patients, of which 93% were low-grade and 7% were high-grade serous carcinoma. Median time to development of subsequent invasive cancer was 10 years, with a range up to 25 years. Presence of a somatic V600E BRAF mutation in patients with serous borderline disease may predict for improved prognosis and decreased likelihood of progression. In a retrospective study by Grisham et al. of 75 tumors of serous borderline or low-grade serous ovarian cancer histology, 57 tumors harbored either a KRAS ( n = 17) or BRAF V600E ( n = 26) mutation. The presence of BRAF V600E mutation was associated significantly with early disease stage (stage I/II; P < 0.001) and serous borderline histology ( P = 0.002). While patients with serous borderline disease are at increased risk of developing low-grade serous ovarian cancer, systemic therapy is reserved for those with confirmed invasive disease.

Fig. 6.2, Development of low-grade serous carcinoma in step wise fashion from.

Pathology

Serous borderline tumors

Gross findings

SBTs can range in size from 1 to 35 cm (mean size: 11.0 cm). The tumors may be purely intracystic, purely surface tumors or involve both cyst wall and the ovarian surface. The tumor grossly is composed of papillary excrescences and the papillae are typically soft, edematous and have a tan-yellow appearance. The cyst contents are mostly serous-type fluid, but can also be mucinous. In some cases, the gross appearance is that of cystadenofibroma and the borderline component is only identified microscopically; therefore, sampling of these areas is important even in the absence of the characteristic papillary excrescences.

Histologic features

Microscopically SBTs demonstrate complex arborizing papillae with hierarchical tufting ( Fig. 6.3 ). By definition there is no evidence of stromal invasion in these tumors, though SBT can be present in the background of low-grade serous carcinoma (LGSC) ( Fig. 6.3 ). The lining epithelium can be either low cuboidal or stratified, and ciliated cells are identified at least focally. Cells with abundant eosinophilic cytoplasm can be seen in a subset of SBTs. Cytologic atypia is mild to moderate, and marked atypia is not a feature of SBTs. Per the WHO criteria, at least 10% of the tumor must show epithelial proliferation to be considered SBT; however, in a few cases, even patients with small foci of SBT can recur. Hence, our practice has been to call these tumors “focal” SBTs if the epithelial proliferation is less than 10%.

Fig. 6.3, Low grade serous carcinoma ( asterisks ) in a background of serous borderline tumor ( arrows ) which demonstrates complex arborizing papillary with hierarchical tufting and cell detachment.

A subset (5%–15%) of SBTs have a micropapillary pattern (SBT-MP). Histologically, these tumors are characterized by large edematous papillae from which emanate slender papillae devoid of fibrovascular cores that are 5 times as long as they are wide, and has been likened to a Medusa head appearance ( Fig. 6.4 ). This pattern must measure ≥ 5.0 mm to meet criteria for a diagnosis of SBT-MP. A subset of SBT-MP can show cribriform architecture, which is thought to result from fusion of the micropapillae. Of note, the term noninvasive LGSC is not accepted for these tumors in the WHO classification of Tumors of Female Reproductive Organs, and is not used at our institution.

Fig. 6.4, Serous borderline tumor with micropapillary pattern characterized by a large edematous papilla from which emanate slender papillae devoid of fibrovascular cores that are 5 times as long as they are wide.

Autoimplants may present either on the surface or between papillae. Histologically, they are composed of plaque-like lesions with features of a desmoplastic noninvasive implant. They are not shown to be associated with adverse outcome.

Another important feature in SBT is the presence of microinvasion. This is defined as the presence of small papillae or single tumor cells with surrounding clear spaces or “halos” measuring < 5 mm in one linear dimension, per the WHO criteria. At our institution we use a cut-off of 3 mm. While microinvasion has not been consistently associated with adverse outcome, multiple foci should prompt additional sampling to exclude microscopic foci of LGSC.

Peritoneal implants associated with SBT are either noninvasive or invasive. Noninvasive implants are divided into epithelial implants and desmoplastic implants, and are collectively called implants. Invasive implants are now considered to be synonymous with LGSC. Epithelial implants are composed of papillary proliferations within cystically lined spaces that are present within the invagination of the omental fat. Desmoplastic implants are characterized by loose spindle cells within which glands, papillae and single cells are embedded, but overall has to be a stroma rich proliferation. When small biopsies are taken without surrounding adipose tissue, distinguishing noninvasive from invasive implants may be difficult, and the implants may be considered indeterminate.

Ancillary testing

SBTs are positive for PAX-8, WT-1, ER, and PR. They typically show patchy staining for p16 and wild-type staining for p53.

Differential diagnosis

The differential diagnosis for SBTs includes high-grade serous carcinoma with a “borderline-like” architectural pattern, but the latter is identified by the presence of marked cytologic atypia. The diagnosis can be further supported by p53 stain which is only overexpressed in high-grade serous carcinoma. Ovarian clear cell carcinoma is another differential, as a subset of SBTs can show cytoplasmic clearing. Presence of solid, and tubulocystic patterns, hobnailing, and marked cytologic atypia would support a diagnosis of ovarian clear cell carcinoma. If needed immunostains such as WT-1 for SBTs, and napsin-A and HNF-1B for ovarian clear cell carcinoma may be considered.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here