ORIGINAL STUDY

Villoglandular Papillary Adenocarcinoma of the Cervix A Series of 28 Cases Including Two With Lymph Node Metastasis Isam M. Lataifeh, MD, CGO,*Þ Maysa Al-Hussaini, MD,þ Catherine Uzan, MD,§ Imad Jaradat, MD,|| Pierre Duvillard, MD,¶ and Philippe Morice, MD#**

Objective: To investigate the clinicopathologic features, the management, and the outcome of villoglandular papillary adenocarcinoma (VGPA) of the uterine cervix. Methods: A retrospective review of patients’ clinical characteristics, pathology, and the disease management, together with outcome information. Results: A total of 28 patients with VGPA were treated. The median age of the patients was 38 years with a range of 26 to 65 years. Sixteen of the 21 patients presented with abnormal bleeding, and 5 patients had an abnormal Papanicolaou (Pap) test result. Nineteen patients had International Federation of Gynecology and Obstetrics stage IB disease, and 5 patients had stage IIB disease. Two of 24 patients, where the lymph node status was known, had positive nodes. Twenty patients underwent different types of radical surgery with or without pelvic radiotherapy, and 8 patients received platinum-based chemotherapy and pelvic radiotherapy with no surgery. The follow-up ranged from 5 to 168 months with a median of 35 months. TwentyYone patients are alive with no evidence of recurrent disease, 5 patients have died because of the disease recurrence, and 2 patients were lost to follow-up. The overall and disease-free 5-year survival for these patients was 82% and 75%, respectively. Conclusion: This study confirms the excellent prognosis of VGPA overall compared to the common forms of cervical cancer, but the prognosis is related to stage and pathology. A large multicenter prospective study is warranted to determine the most appropriate treatment for the disease. Until then, a meta-analysis on the subject would be of benefit. Key Words: Adenocarcinoma, Cervix, Papillary, Villoglandular, Lymph node Received January 22, 2013, and in revised form February 23, 2013. Accepted for publication February 25, 2013. (Int J Gynecol Cancer 2013;00: 00Y00)

papillary adenocarcinoma (VGPA) of the V illoglandular uterine cervix was described for the first time by Young

and is distinguished from the common types of adenocarcinoma by its generally good long-term prognosis,2,3 as it infrequently involves lymph nodes (LNs).4 Macroscopically, those tumors are exophytic. Villoglandular papillary adenocarcinomas were superficially infiltrative tumors with

*Department of Obstetrics and Gynecology, Jordan University of Science and Technology, Irbid, Jordan; Departments of †Surgery and ‡Pathology, King Hussein Cancer Center, Amman, Jordan; §Department of Surgery, Institute Gustave Roussy, Paris, France; ||Radiation Therapy, King Hussein Cancer Center, Amman, Jordan; ¶Department

of Pathology, Institute Gustave Roussy, Paris in France; and #Department of Surgery, Institute Gustave Roussy, Paris, France; Unit INSERM 10-30, Villejuif, France; and **University Paris Sud, Le Kremlin Biceˆtre, France. Address correspondence and reprint requests to Isam M. Lataifeh, MD, CGO, Department of Obstetrics and Gynecology, Jordan University of Science and Technology, PO Box 3030, Irbid 22110, Jordan. E-mail: [email protected]. The authors declare no conflicts of interest.

and Scully in 1989.1 Villoglandular papillary adenocarcinoma is an uncommon variant of well-differentiated endocervical adenocarcinoma.1 It usually affects young women

Copyright * 2013 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0b013e31828efcaa International Journal of Gynecological Cancer

& Volume 00, Number 00, Month 2013

Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

1

Lataifeh et al

International Journal of Gynecological Cancer

exophytic architecture, but these tumors were cytologically low-grade lesions, with a low number of mitosis, minimal cytologic atypia, and minimal vascular space invasion. Because of the young age of patients and the favorable prognosis, conservative surgery such as cone biopsy is adopted by many physicians when the tumor is low grade and superficial, without vascular space invasion or involvement of the resection margins.2,5Y8 Accurate pathological recognition of the tumor entity and differentiation from other forms of cervical adenocarcinoma is essential for the safety of conservative surgical approaches. In this study, we reviewed the clinical and pathological features of this neoplasm at 2 institutions: King Hussein Cancer Center (KHCC) in Jordan and Institute Gustave Roussy (IGR) in France.

& Volume 00, Number 00, Month 2013

performed using SAS version 9.2 (SAS institute Inc, Cary, NC). Ethical approval from the institutional review boards of the 2 institutions (KHCC and IGR) was obtained.

RESULTS Clinical Characteristics, Staging, and Pathologic Findings During the study period, a total of 28 patients with a diagnosis of VGPA were treated at the 2 centers. The distribution by center was 22 patients (78.6%) at IGR and 6 patients (21.4%) at KHCC. The distribution of patients according to the clinical characteristics is shown in Table 1. The details of the patients’ characteristics, pathology of tumor, and management of the disease are shown in Table 2.

MATERIALS AND METHODS A retrospective review of all cases of invasive adenocarcinoma of the cervix identified from the surgical pathology electronic files at the KHCC in Jordan between January 2003 and December 2011 and the IGR, Villejuif, France, between January 1995 and December 2011 was conducted. The medical records of all patients with the histological diagnosis of VGPA were reviewed to collect data on the patients’ characteristics, pathology of the tumor, and the management of the disease. The patients’ characteristics included age, marital status, parity, presenting symptoms, smoking, oral contraceptive pills (OCP) use, and Pap test history. Information on the pathology of the tumor included stage, grade, LN status, method of diagnosis, and microscopic pathologic findings. Data on the management of the disease included type of primary treatment, follow-up, disease recurrence, site of recurrence, and disease-free and overall survival. The pathologic diagnosis was made based on analysis of biopsies or cone specimens. All slides had been reviewed by an expert pathologist (M.A. for the KHCC and P.D. for the IGR). The diagnosis of VGPA was accepted if papilla formation with a predominantly exophytic growth pattern accounted for more than 50% of the tumor. The presence of endophytic invasive component with papillary, tubular, microglandular, or mixed patterns was accepted if accounted for more than 50% of the tumor. The papilla cores could be thin or thick, if estimated to be less or more than 1 cell thickness, respectively, with or without inflammation with endocervical-like or desmoplastic stroma. The lining epithelium of endocervical, endometrial, intestinal, or mixed types is accepted. Potential differential diagnosis is endocervical adenocarcinoma that is conventionally more infiltrative with severe atypia, a higher number of mitosis, and with a higher grade than VGPA. Other types of adenocarcinoma (endometrioid, clear cell, and serous papillary) were excluded. The patients’ clinical characteristics, International Federation of Gynecology and Obstetrics (FIGO) stage distribution, and LN status were presented as counts and percentages. Descriptive statistics (mean, median, and minimum and maximum) were calculated for age and follow-up. All analyses were

2

TABLE 1. Distribution of patients according to clinical characteristics Clinical Characteristics

n (%)

Age, yrs 20Y50 950 Smoke Unknown No Yes Presenting symptom Unknown Asymptomatic Abnormal vaginal bleeding Postmenopausal bleeding Abnormal bleeding during pregnancy Diagnosed during clinical examination Pelvic pain FIGO stage IA IB1 IB2 IIA IIB IIIA IIIB IVA IVB LN status Histologic nodal status lymphadenectomy not done (imaging staging) Negative Positive

20 (71.4) 8 (28.6) 13 9 (60) 6 (40)

5 12 1 3 1 1

5 (21.7) (52.2) (4.3) (12.9) (4.3) (4.3)

0 (39.3) (28.6) (7.2) (17.9) 0 2 (7.2) 0 0

11 8 2 5

7 19 (92) 2 (8)

* 2013 IGCS and ESGO

Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

26

56 65 61

61

44

37

43 39 32

55 57

37

28

31

31

1

2 3 4

5

6

7

8 9 10

11 12

13

14

15

16

Age No (Yrs)

* 2013 IGCS and ESGO

Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

Cervical mass during pregnancy Punch Bx, VGPA Cervical mass, punch Bx, VGPA

Cervical mass, punch Bx, VGPA

Cervical mass, punch Bx, VGPA

Cervical mass during pregnancy, punch Bx, VGPA Cervical mass, punch Bx, VGPA Cervical mass, cone Bx, VGPA Cervical polyp during pregnancy, punch Bx, VGPA Cervical mass, punch Bx, VGPA Cervical mass, punch Bx, VGPA

Cervical mass, punch Bx, VGPA

Cervical mass, punch Bx, VGPA

Cervical mass, punch Bx, VGPA Cervical mass, punch Bx, VGPA Cervical mass, punch Bx, VGPA

Cervical mass, punch Bx, VGPA

Reason for Referral

j PL

j PL

j (PET/CT)

j (PET/CT)

j PET/CT j PA

j PL j PL j PL

j PA

j PL

+ PL/PA

LN Status

33

31

108

35

5 12

18 35 6

11 Lost to follow-up Lost to follow-up 26 43 6 12 8 10 6

8

Follow-up (Months)

C/s, lap staging, chemo/RTX and brachy Brachy and lap RH, PLND Lap RH and PLND C/s and cone Bx, radical trachel and PLND Chemo/RTX and brachy Lap staging, chemo/RTX and brachy Lap staging, chemo/RTX and brachy then completion lap hys Chemo/RTX and brachy then TAHPAND C/s and PAND, chemo/RTX and brachy Brachy and lap RH and PLND

Chemo/RTX

Chemo/RTX and brachy

RH, BSO, LND Chemo/RTX and brachy Chemo/RTX and brachy

PL/PAND, chemo

Treatment

NED

NED

NED

NED

NED NED

NED NED NED

ROD DOD ROD DOD Persistent disease DOD NED

Persistent disease and RTX/ Brachy DOD

Outcome

& Volume 00, Number 00, Month 2013

IB1 (30 mm)

IB2 unknown

IB2 (50 mm)

IIB

IIB IB2 (45 mm)

IB1 (35Y40 mm) IB1 (12 m) During IB1 (35 mm)

IB2 (45 mm)

IIIB

IIB

IIA IIA IIB

IIIB

FIGO Stage

TABLE 2. Details of patients’ characteristics, pathology of tumor, and management of the disease

International Journal of Gynecological Cancer Cervical VGPA

3

4

49

29

27

30

28 56

57

37

31

40

19

20

21

22

23 24

25

26

27

28

Cervical mass, punch Bx, VGPA Mixed with endometr, breast mass, Bx adenocarcinoma Abnormal smear result, cone Bx, VGPA Cervical mass, punch Bx, VGPA Mixed with endometr

Cervical mass, punch Bx, VGPA

Abnormal smear result, cone Bx, VGPA Cervical mass, punch Bx, VGPA Abnormal smear result, cone Bx, VGPA

Abnormal EC, curettage and cone Bx, VGPA Abnormal smear result, cone Bx, VGPA

Cervical lesion, punch Bx, VGPA Abnormal smear result, cone Bx, VGPA Cervical mass, punch Bx, VGPA

Reason for Referral

IIB

IB1 (10 mm)

IB2 (70 mm)

IB2 (45 mm)

IB2 (70 mm) IB2 (45 mm)

IB1 (15 mm)

IB1 (25 mm)

IB1 (30 mm)

IB1 (10 mm)

IB1 (25 mm) IB1 (12 mm)

FIGO Stage

Chemo/RTX, brachy, completion TAH and PLND RH, PL/PALND, brachy for cervix, mastectomy, ALND RTX to breast Rad trachel PLND Chemo/RTX, brachy completion TAH and PALND

j (PET/CT)

j PA

j PL

j PL

DOD NED

36 108

97

60

168

NED

NED

NED

NED

NED

NED

NED

NED NED

Outcome

NED ROD

54

19

54

120

41 30

Follow-up (Months)

80 24

Brachy and lap RH and PLND Rad trachel and PLND

Treatment

j PA and PL Brachy and RH and PA and PLND j PL Brachy and lap RH and PLND j sentinel Brachy and lap RH and PLND j PL + PL Lap RH and PLND chemo/RTX (extended field) j (PET/CT) Chemo/RTX and brachy j (PET/CT) RTX alone, medically unfit

j PL j PL

LN Status

International Journal of Gynecological Cancer

ALND, axillary lymph nodes dissection; brachy, brachytherapy; BSO, bilateral salpingo-oophorectomy; Bx, biopsy; chemo, chemotherapy; C/s, cesarean section; CT, computed tomography; DOD, death from disease; endometr, endometrium; hys, hysterectomy; lap, laparoscopic; LVSI, lymphovascular space invasion; NED, no disease detected; PA, para-aortic; PET, positron emission tomography; PL, pelvic; RH, radical hysterectomy; ROD, recurrence of disease; RTX, radiotherapy; TAH, total abdominal hysterectomy; trachy, trachelectomy; j, negative; +, positive.

46 32

17 18

Age No (Yrs)

TABLE 2. (Continued)

Lataifeh et al

& Volume 00, Number 00, Month 2013

* 2013 IGCS and ESGO

Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

International Journal of Gynecological Cancer

& Volume 00, Number 00, Month 2013

The median age of the patients was 38 years with a range of 26 to 65 years (20 patients were younger than 50 years). The most common symptom was abnormal bleeding in premenopausal or postmenopausal women (13 patients). Three patients had their tumor diagnosed during pregnancy (Table 1). Nineteen patients (68%) had FIGO stage IB disease, 7 patients had FIGO stage II disease, and 2 patients had stage IIIB disease. The LN status was determined on conventional imaging alone (abdominopelvic magnetic resonance imaging and/or computed tomographic [CT] scan) in 4 patients by using conventional imaging and positron emission tomography (PEC) and CT (PET/CT) in 3 patients. Twenty-one patients had LN surgical staging (3 of them having a previous PET/CT imaging), and 2 patients had nodal involvement (cases 1 and 22 in Table 2). One patient had stage IIIB disease with multiple bulky pelvic and para-aortic LNs (case 1), and the other one had stage IB1 (15 mm) disease with a single pelvic metastasis out of 22 LN lymph nodes removed (case 22).

Treatment In stage IB1 disease (11 patients), all patients were treated by upfront surgery including radical hysterectomy or trachelectomy and pelvic lymphadenectomy (5 patients) or a combination of preoperative brachytherapy followed by radical hysterectomy and dissection of pelvic nodes (6 patients) (Table 2). Among 5 patients having an upfront surgery, 3 patients had radical trachelectomy (after frozen section analysis of pelvic LNs) and 2 patients had a radical hysterectomy. One of these 2 patients had pelvic node involvement (case 22) and underwent postoperative chemoradiation therapy. Among 8 patients with stage IB2 disease, 6 patients were treated by chemoradiation therapy (with concurrent cisplatin-based chemotherapy) and uterovaginal brachytherapy (with 3 patients having a pretherapeutic para-aortic staging laparoscopic procedure). One patient with a poor medical status was treated by external radiation without concurrent chemotherapy (case 24). The last patient with stage IB2 disease was treated using upfront surgery followed by postoperative brachytherapy. Among 9 patients with disease stage more advanced than stage II, all but one were treated by chemoradiation therapy and a uterovaginal brachytherapy (one of them with pelvic and para-aortic nodal involvement having an extended field of radiation therapy/case 1). Two of these patients underwent initial para-aortic lymphadenectomy. One patient with stage IIA disease underwent an exclusive surgery (case 2).

Follow-up and Outcomes The follow-up ranged from 5 to 168 months with a median of 35 months. Twenty-one patients (75%) were alive with no evidence of recurrent disease at the time of review, 5 patients (18%) had died of disease recurrence, and 2 patients (7.1%) were lost to follow-up. The details of the patients who died from disease recurrence are as follows: 2 patients presented with locally advanced-stage disease (stage IIIB), with one of them having metastasis to the pelvic and para-aortic lymph nodes. Both had persistent disease after chemotherapy and extended field

Cervical VGPA

pelvic radiotherapy and had died at 11 and 10 months after diagnosis. Another 2 patients presented with stage IIB disease and received chemotherapy and pelvic radiotherapy. They developed recurrence of disease in the cervix 6 and 26 months after diagnosis and died 12 and 43 months, respectively, after diagnosis. The fifth patient presented with stage IB2 disease with negative nodes and received pelvic radiotherapy alone because of medical comorbidities. She developed peritoneal recurrence 24 months after diagnosis and died 36 months after diagnosis. The actuarial overall and disease-free 5-year survival for these patients were 82% and 75%, respectively.

DISCUSSION Clinical Characteristics Villoglandular papillary adenocarcinoma has recently been recognized and identified as a distinctive histological entity from the usual invasive adenocarcinoma of the uterine cervix.1 It has been reported to account for approximately 4% of cervical adenocarcinomas in the English literature.5 Villoglandular papillary adenocarcinoma is generally found in younger women than squamous cell carcinoma. In this series, the median age of patients was 38 years with a range of 26 to 65 years, with 71.4% of the patients between the ages of 26 and 50 years. This is similar to the reported series in the literature2,9,10 that reported a total of 52 cases, with patients’ ages ranging between 22 and 64 years and the average age of onset between the age of 37 and 45 years. Bulk et al11 reported that the median age of patients with adenocarcinoma was 3 years below the median age of patients with squamous cell carcinoma, that is, 44 years (range, 18Y92 years) and 47 years (range, 19Y98 years), respectively. The association of OCP use with the development of adenocarcinoma is controversial. Dallenbach-Hellweg12 reported that 82% of women with adenocarcinoma of the cervix had a history of OCP use, compared to 40% of women with squamous cell carcinoma. Jones et al2 reported that 15 (62.5%) of 24 patients had taken OCP, with a duration of use (when known) ranging from 5 to 20 years. Young and Scully1 reported that only 1 of 13 patients had a history of OCP use. In the present study, the data on OCP use was not available for most of the patients. Meanwhile, high-risk types of human papillomavirus have been detected in almost all cervical cancers.13 However, to the best of our knowledge, there is no report about the relationship of this condition and human papillomavirus infection. Villoglandular papillary adenocarcinoma is seen in much the same group of women as squamous cell carcinoma and adenocarcinoma. It presents in much the same way as other cervical cancers and is only really recognized as different by biopsy. An excisional or cone biopsy is sometimes required to establish the diagnosis of VGPA when a punch biopsy is not enough.

Prognostic Factors and Implications for the Treatment The incidence of LN metastasis was low in this series; 1 (8.3%) of 12 patients with stage IB1 disease had positive

* 2013 IGCS and ESGO

Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

5

Lataifeh et al

International Journal of Gynecological Cancer

LNs. The metastasis to the LN was surgically determined in 2 patients who had stage IIIB and IB1 diseases, respectively. A similar low incidence was noted in the series of Khunamornpong et al9 and Utsugi et al10 who reported positive LNs in 2 of 15 patients and 1 of 13 patients, respectively. A more favorable outcome for VGPA than for the common types of cervical adenocarcinoma has been reported in several series.1,2 In reports in the English literature,1,2,5,9,10 only 1 of 72 patients with VGPA developed recurrence. Kaku et al5 reported a patient who had stage IIB disease and who underwent radical hysterectomy and lymphadenectomy followed by radiation therapy but developed vaginal recurrence at 30 months and died of the disease at 46 months, after primary treatment. The previously reported excellent prognosis of this variant of tumor raises the question of a conservative surgical approach, especially in young patients who want to preserve reproductive function; but this approach should be adopted with caution because there are a number of reports of unfavorable outcomes with this type of tumor. The authors acknowledge that prognosis is driven by stage and not just histology alone, as other ‘‘conventional’’ histologic subtypes of cervical cancer (squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinomas). As a matter of fact, in this series, 9 patients had more than stage IB disease and only 5 patients had known recurrence/ death (2 would be censored owing to lack of follow-up). No patient with stage IB disease had a recurrence and died compared to the standard 15% recurrence rate for stage IB SCC disease, suggesting a more favorable prognosis stage for stage. The main limitations of this study were its being retrospective, reporting a relatively small number of patients. The strength of this study is reporting data from 2 referral oncology institutions with long period of follow-up and is being the largest series ever reported on VGPA. This study confirms the excellent prognosis of VGPA overall compared to the common forms of cervical cancer, but the prognosis is related to stage and pathology; and 9 of the 28 patients in this series had a disease more advanced than stage IB. This is in agreement with other reports that confirm that the prognosis of VGPA is good for early-stage disease but not all diseases.14 Published reports on VGPA are sparse. The authors believe that a large multicenter prospective study is warranted to define the best treatment for the disease and to clarify more fully the prognostic factors other than FIGO stage; and even then, local variations of treatment modalities will make

6

& Volume 00, Number 00, Month 2013

comparisons difficult. Until then, a meta-analysis on the subject would be of benefit.

REFERENCES 1. Young RH, Scully RE. Villoglandular adenocarcinoma of the uterine cervix. A clinicopathologic analysis of 13 cases. Cancer. 1989;63:1773Y1779. 2. Jones MW, Silverberg SG, Kurman RJ. Well differentiated villoglandular adenocarcinoma of the uterine cervix: a clinicopathologic study of 24 cases. Int J Gynecol Pathol. 1993;12:1Y7. 3. Stanley-Christian H, Heim BK, Hines JF, et al. Villoglandular adenocarcinoma of the cervix: a report of three cases and review of literature. Gynecol Oncol. 1997;66:327Y330. 4. Macdonald RD, Kirwan J, Hayat K, et al. Villoglandular adenocarcinoma of the cervix: clarity is needed on the histological definition for the difficult diagnosis. Gynecol Oncol. 2006;100:192Y194. 5. Kaku T, Kamura T, Shigematsu T, et al. Adenocarcinoma of the uterine cervix with predominantly villoglandular papillary growth pattern. Gynecol Oncol. 1997;64:147Y152. 6. Costa MJ, McInay KR, Trelford J. Cervical carcinoma with glandular differentiation: histological evaluation predicts disease recurrence in clinical stage I or II patients. Hum Pathol. 1995;26:829Y837. 7. Hopson L, Jones MA, Boyce CR, et al. Papillary villoglandular carcinoma of the cervix. Gynecol Oncol. 1990;38:221Y224. 8. Ballo MS, Silverberg SG, Sidaway MK. Cytologic features of well-differentiated villoglandular adenocarcinoma of the cervix. Acta Cytol. 1996;40:536Y540. 9. Khunamornpong S, Maleemonkol S, Siriaunkgul S, et al. Well-Differentiated villoglandular adenocarcinoma of the uterine cervix: a report of 15 cases including two with lymph node metastasis. J Med Assoc Thai. 2001;84:882Y888. 10. Utsugi K, Shimizu Y, Akiyama F, et al. Clinicopathologic features of villoglandular papillary adenocarcinoma of the uterine cervix. Gynecol Oncol. 2004;92:64Y70. 11. Bulk S, Visser O, Rozendaal L, et al. Incidence and survival rate of women with cervical cancer in the Greater Amsterdam area. Br J Cancer. 2003;89:834Y839. 12. Dallenbach-Hellweg G. On the origin and histology structure of adenocarcinoma of the endocervix the endocervix in women under 5 years of age. Pathol Res Pract. 1984;179:38Y50. 13. Kleter B, van Doorn LJ, ter Schegget J, et al. Novel short-fragment PCR assay for highly sensitive broad-spectrum detection of anogenital human papillomaviruses. Am J Pathol. 1998;153:1731Y1739. 14. Kong CS, Longacre TA, Hendrickson MR. Pathology. In: Berek J, Hacker N, eds. Gynecologic Oncology. 5th ed. Philadelphia, PA: Lippincott Williams &Wilkins; 2010:137.

* 2013 IGCS and ESGO

Copyright © 2013 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.