Early-Stage, BRCA-Associated Ovarian Cancer Detected by Papanicolaou Smear: A Case Report

Historically known as a "silent killer", ovarian cancer is often diagnosed at an advanced stage. We describe an unusual case of stage I, ovarian, high-grade serous carcinoma detected by a routine Papanicolaou (PAP) smear, with no abnormal physical, imaging, or laboratory findings. A 53-year-old woman with newly diagnosed triple-negative breast cancer received a screening Pap smear, which showed malignant cells not coming from the breast or uterine cervix. Pelvic examination, cervical biopsy, and gynecologic ultrasonography found no abnormality. Endometrial curettage yielded free-floating adenocarcinoma cells. The immunohistochemical stain result indicated ovary or fallopian tube cancer. Complete cytoreductive surgery was performed, and high-grade serous carcinoma of bilateral ovaries, FIGO stage IB, was diagnosed. Although extremely rare, when malignant cells not originating from the uterine cervix are detected on a Pap smear, it may lead to an early diagnosis of ovarian cancers, and this warrants further comprehensive workup.


Introduction
Ovarian cancer is historically called a "silent killer." Women with ovarian cancer are often diagnosed at an advanced stage. The overall stages at the diagnosis of ovarian cancer are 23% in stage I, 8% in stage II, 34% in stage III, and 26% in stage IV [1]. There is still no efficient screening method for early ovarian cancer detection in prospective randomized controlled trials [1]. Most patients with ovarian cancer present with nonspecific symptoms. The most common symptom presenting in early-stage ovarian cancer is abdominal or pelvic pain [2]. The most common symptom in an advanced stage is abdominal swelling caused by ascites. Some women experience other non-specific symptoms, such as pelvic or abdominal pain, indigestion, and altered bowel habits, before diagnosis [1].
The Papanicolaou (Pap) smear has been widely used to screen for precancerous and malignant cervical diseases. In 1946, Papanicolaou purposed the diagnostic value of a cervicovaginal smear for its ability to collect exfoliated cancer cells from the lower genital tract [3].
The uterine cervix is rarely metastasized by ovarian cancers. First, the lymphatic drainage system directs cancer cells away from the cervix. Secondly, the cervix consists mainly of fibrous tissue, which is not favored by ovarian cancer cells to seed on [4]. Nevertheless, exfoliated ovarian cancer cells may travel through the fallopian tubes and reach the endocervix, hence they are occasionally collected by a cervicovaginal smear. The detection of ovarian cancer cells by a cervicovaginal smear usually indicates an advanced clinical stage, as reported in the literature [5][6][7].
Here, we present a case with early-stage, high-grade serous ovarian malignancy detected by a screening Pap smear while she received neoadjuvant chemotherapy for her breast cancer.

Case Presentation
Case presentation, physical, imaging, and laboratory findings A 53-year-old Taiwanese woman presented with a right breast lump for more than one month. Before that, her only medical record was cervical condyloma treated by excision. She received a breast biopsy and was diagnosed with triple-negative, clinical stage Ic invasive ductal carcinoma of her right breast in a local hospital. While she was having neoadjuvant chemotherapy with Doxorubicin, Cyclophosphamide, and Bevacizumab, she received a screening Pap smear. Malignant cells, not squamous cell carcinoma or adenocarcinoma, were identified.

Tissue biopsy evaluation
A random cervical biopsy revealed chronic cervicitis. Endocervical curettage found normal glandular cells. Endometrial curettage yielded proliferative endometrium with a few free-floating adenocarcinoma cells and normal endometrial glands. Therefore, the nature of the malignant cells was still undetermined. We performed immunohistochemical (IHC) stains. These tumor cells were positive for Wilms' tumor 1 (WT1), Paired-box gene 8 (PAX8), p53, and p16, and they were negative for estrogen receptor (ER), progesterone receptor (PR), gross cystic disease fluid protein 15 (GCDFP-15), and GATA binding protein 3 (GATA3). We compared the previous IHC findings of breast specimens done at another hospital, which were negative for ER, PR, WT1, and PAX8. The IHC findings suggested that these free-floating cells, collected from the Pap smear, were more likely to be originated from the ovary or fallopian tube, rather than from metastasis of her breast cancer ( Figure 2).

Surgery and pathologic findings
The patient underwent complete cytoreductive surgery, including total hysterectomy, bilateral salpingooophorectomy, pelvic lymph node dissection, para-aortic dissection, and omentectomy. The uterus, bilateral adnexa, and other pelvic organs appeared normal by the surgeon's inspection and palpation ( Figure 3).

FIGURE 3: The resected specimen
The specimen consists of a resected uterus, bilateral fallopian tubes, and ovaries. Grossly, there are only several myomas. No other tumor or other abnormal structure is seen in the endometrial cavity and myometrium.
The patient had no obvious ascites, and peritoneal washing cytology was negative for malignancy cells. The pathology, however, revealed high-grade serous carcinoma of bilateral ovaries ( Figure 4, Figure 5).

FIGURE 4: The histologic examination of the left ovary
The largest tumor lesion site in the left ovary. It showed poor differentiated serous carcinoma (Hematoxylin and Eosin, x30 magnification). The size of the lesion is labeled.

FIGURE 5: The histologic examination of the right ovary
Only some single cancer cells were found in the right ovary.

Arrows: carcinoma cells
The endometrium was thoroughly sampled for a slide review (all for section) and no atypical cells or malignant cells were found. The bilateral fallopian tubes were sectioned following the sectioning and extensively examining the fimbriated end (SEE-FIM) protocol, and no atypical cells or malignant cells were found ( Figure 6). No masses or tumors were observed on the gross specimen of the omentum by inspection and palpation. The omentum was sectioned with one-centimeter intervals. The three most representative sections were selected for microscopic examination, and no atypical cells or malignant cells were found. The uterine serosa, myometrium, cervix, parametrium, pelvic lymph nodes, and para-aortic lymph nodes were all free of tumors. We performed WT1, PAX8, GCDFP-15, GATA-3, special AT-rich sequence-binding protein 2 (SATB2), and SRY-box transcription factor 10 (SOX10) on the specimen of ovaries. The WT1 and PAX8 were positive while the GATA-3, SATB2, and SOX10 were negative. We checked the p53 signature of bilateral fallopian tubes, and the result was the wild-type pattern. Therefore, the cancer was considered a new primary ovarian cancer. Theoretically, cancer cells could have been present in the ascites at some moment, and the stage should be IC, but they were not detected in the final peritoneal washing cytology. According to the above objective pathologic findings, the FIGO stage was IB.

Postoperative treatment and follow-up
After the surgery, the patient received four cycles of chemotherapy with Docetaxel and Carboplatin, followed by a bilateral total mastectomy. The pathologic report revealed stage I triple-negative breast cancer. The patient had regular follow-ups for both ovarian cancer and breast cancer. Till the time this article was drafted, four years after the treatment, there was no sign of any tumor recurrence.
During her follow-ups, more of her family history was obtained. Her mother and grandmother died of endometrial cancer many years ago. Her sister was also diagnosed with triple-negative breast cancer six months after the patient had finished her treatment. Due to multiple BRCA-related cancer family histories, the patient and her family members received genetic testing. The patient, her sister, and her daughter were found to have the BRCA 1 c.212 C>A mutation.

Discussion
Epithelial ovarian cancer is one of the leading causes of death in gynecologic malignancies. In its early stage, affected women generally have no or few specific symptoms. Therefore, the disease is commonly detected at an advanced stage. Epithelial ovarian cancers are classified into five main types: high-grade serous, low-grade serous, endometrioid, clear cell, and mucinous [1]. The most prevalent histological type is high-grade serous carcinoma, which is usually diagnosed at FIGO stage III (51%) or IV (29%) [1]. Currently, there is still no reliable screening method to detect early-stage ovarian cancer. Universal screening with transvaginal ultrasound alone or in combination with CA-125 did not decrease mortality from ovarian cancer in average-risk women [8]. Major cancer societies, including the U.S. Preventative Services Task Force, the American Cancer Society, the American Congress of Obstetricians and Gynecologists, and the National Comprehensive Cancer Network (NCCN), all recommend against routine screening for all populations.
For high-risk women, especially those carrying a BRCA gene mutation, the approach is very different. The lifetime risk of ovarian cancer is 39% to 59% in BRCA 1 gene mutation carriers and 11% to 20% in BRCA 2 gene mutation carriers [9]. The NCCN suggested comprehensive counseling and considering risk-reducing salpingo-oophorectomy (RRSO), favorably between 35 and 40 years old, if childbearing is completed. For those women not receiving RRSO, despite no proven benefit, yearly transvaginal ultrasound combined with serum CA-125 testing starting from age 30 can be considered [10].
Ovarian cancer cells spread mainly through the exfoliation of cells along the peritoneal surface or lymphatic dissemination, and hematogenous metastasis is less common [8]. Daw et al. purposed that the high content of fibrous tissue in the cervix, the small size of the cervix, and the lymphatic drainage of the pelvis in a direction away from the cervix may be the reasons why the cervix is unfavorable to metastasis [4]. In our case, we suspected that in rare circumstances, ovarian tumor cells drop on fimbria, travel through the fallopian tubes and the uterine cavity, and reach the surface of the uterine cervix without seeding into any organ on the path.
After George Papanicolaou first published his work on the use and diagnostic procedure of vaginal smear for uterine cancer [11], he and his colleagues published another article to describe the diagnostic value of a cervicovaginal smear for its ability to collect exfoliated cancer cells from the lower genital tract or even other intra-pelvic organs like the urinary bladder or kidneys [3]. Later articles suggested that in a routine Pap smear, the detection of free-floating tumor cells, psammoma bodies, or atypical glandular cells (AGC) on cervical cytology may indicate malignancies other than those of cervical origin [5]. The free-floating tumor cells or other abnormal cells detected by Pap smear may come from tumor direct extension if the tumor was adjacent to the cervix; along with ascites flow if there was ascites formation or through patent fallopian tubes.
Mulvany et al. stated in a review article that there are various ways for tumor cells to spread to the cervix, and cancers of the endometrium, fallopian tubes, ovaries, or even extra genital organs all have the possibility to present with malignant cells in a Pap smear [12]. Those findings often indicate disseminated diseases. To distinguish the real origin, a thorough review of specific diagnostic features of each suspicious organ is necessary.
From our literature review, non-cervical cancers presenting with abnormal Pap smear findings were not rare. However, in almost all articles, the cancers were diagnosed at a late stage, i.e., stage III or stage IV. A sixyear follow-up study reported that the positive predictive value of malignant cytology by the Pap smear for all kinds of gynecological or non-gynecological malignancies was 89.7%, with endometrial carcinoma being the most frequent diagnosis [13]. In another article on abnormal Pap smear findings in extrauterine malignancies, the ovary, fallopian tubes, and gastrointestinal tract were the three most common origins [7]. Nwanodi et al. stated in a review that 10% to 40% of ovarian cancer demonstrated malignant cells on cervicovaginal cytology [5]; nine out of 11 were at grade III or a more advanced stage. Tepeoğlu et al. reviewed six cases of ovarian serous borderline tumors presented as atypical glandular cells and psammoma bodies in a Pap smear [6], and none were diagnosed at a FIGO stage less than IIB. We summarize similar cases from the literature in Table 1. To the best of our knowledge, our case is the first reported case in which stage IB ovarian cancer was detected by a Pap smear.

Cervicovaginal cytology Diagnosis Stage
Athavale   [15]. Kinde et al. reported the detection of some commonly mutated genes associated with endometrial cancers and ovarian cancers by analyzing the fluid collected from a standard liquid-base Pap smear, with a sensitivity of 100% for endometrial cancer and 41% for ovarian cancer [16].
Other techniques, such as analyzing DNA methylation in certain genes extracted from scraped cervical cells, also showed promising results for the early detection of endometrial and ovarian cancers [17]. However, with many techniques mentioned and developed, there is still not a single screening method that can be effective enough to reduce the mortality of high-grade serous ovarian cancer.

Conclusions
Pap smear, a widely accepted screening test for cervical cancer, has the potential to detect ovarian cancer, though most reported cases were at a later clinical stage. We present a case with early-stage ovarian cancer detected by a screening Pap smear, with an absence of any abnormal findings by physical examination, serum tumor markers, pelvic ultrasound, and abdominal CT. It reminds us that although rare, an abnormal Pap smear may lead to the early diagnosis of non-cervical cancers, and it warrants further comprehensive workup. With the development of new methodologies, the early detection of tumor cells or DNAs from ovarian cancer by this relatively simple exam seems achievable and more studies regarding the feasibility and cost-effectiveness of new techniques are necessary.