Share this article on:

The Presence of Human Papillomavirus or p16 in Six Cases of Retroperitoneal Carcinoma

Clements, Aine MD; Euscher, Elizabeth MD; Lacour, Robin MD; Merritt, William MD; Klopp, Ann MD; Ramondetta, Lois MD

doi: 10.1097/AOG.0b013e3181f88ddf
Original Research

OBJECTIVE: To describe six cases of unknown primary carcinoma of the retroperitoneum that were positive for human papillomavirus (HPV), a surrogate molecular marker for high-risk HPV (p16), or both.

METHODS: Using the MD Anderson pathology database, females with confirmed HPV or p16 expression within retroperitoneal carcinomas of unknown primary were identified. Clinical data were collected by retrospective chart review. One pathologist reviewed all histology. Individuals with known primary were excluded. Data regarding individual demographics, presentation, Pap test history, pathology, HPV and p16 positivity, and outcome were analyzed using descriptive statistics.

RESULTS: Six individuals were identified. The median age of the individuals was 43.5 years (range 27-54). Three malignancies (50%) were squamous and three (50%) were undifferentiated. Median follow-up was 12 months (range 6-48 months). Two of the six (33%) individuals had remote histories (more than 10 years) of abnormal Pap test results. All had normal gynecologic examination and Pap test results at diagnosis. Four tumors were HPV-positive (66%) and six were p16-positive (100%). All samples that were HPV-positive were also p16-positive. All individuals underwent treatment with chemotherapy, radiation, or both. One individual underwent initial attempt at resection that was unsuccessful. Two individuals are without evidence of disease, two have had progression of disease, and two have died of their disease.

CONCLUSION: Pelvic masses of unknown primary may be HPV-related despite normal cervical examinations.


Six retroperitoneal carcinomas of unknown primary were found to be positive for human papillomavirus, p16, or both.

From the Women and Infants Hospital/The Warren Alpert Medical School, Brown University, Providence, Rhode Island; the Departments of Pathology, Gynecologic Oncology, and Radiation Oncology, University of Texas, MD Anderson Cancer Center, Houston, Texas; and South Carolina Oncology Associates, Columbia, South Carolina.

Corresponding author: Aine Clements, University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, IA 52242; e-mail:

Financial Disclosure The authors did not report any potential conflicts of interest.

Human papillomaviruses (HPV) were first identified as carcinogens in the 1980s. Since then, p16 has been identified as an immunohistochemical tool to help identify the presence of HPV.1,2 HPV combined with p16 testing has been shown to increase the sensitivity of detection of cervical intraepithelial neoplasia.3 Often, p16 testing is performed on malignant tissue if HPV infection is suspected. Many malignancies, both gynecologic and nongynecologic, such as oropharyngeal, prostate, bladder, breast, esophageal, and lung, have been identified as containing HPV.4

Human papillomavirus is thought to infect cells through direct contact. Human papillomavirus causes loss of control of cell cycle differentiation. Furthermore, E6 and E7, two proteins encoded for by HPV, cause genomic instability. The combination of loss of cell cycle differentiation and genomic instability is believed to be what contributes to the development of a carcinoma.4

Human papillomavirus is important in the development of malignancies of the gynecologic system. Ninety-five percent of cases of cervical cancer are HPV-positive. Approximately 50% of vulvar squamous cell cancers and 64% to 91% of vaginal cancers are HPV-positive. Human papillomavirus is only rarely positive in endometrial cancers or ovarian cancers.5–8 This distribution of HPV infection is likely related to the increased infectiousness with direct contact to an HPV vector.

However, HPV has also been identified in nongynecologic cancers such as oropharyngeal, prostate, bladder, breast, esophageal, and lung cancers.1 HPV has not been previously described in retroperitoneal pelvic malignancies of unknown primary. We describe six cases of women who presented with retroperitoneal pelvic malignancies of unknown primary that were HPV-positive or p16-positive. We present their clinical course and treatment outcomes.

Back to Top | Article Outline


Approval was obtained by the MD Anderson Cancer Center institutional review board. Review of the pathology database at MD Anderson Cancer Center, Houston, Texas, was performed from the years 2000 to 2009. The search terms “retroperitoneal” and “squamous” were entered in the database through two separate searches. After each search term, individuals were excluded if the primary site was known. Each individual chart was then examined to see if HPV testing and p16 testing were performed. If not, then stored tissue was tested, if possible. Clinical data of the individuals who had HPV-positive or p16-positive tumors were collected retrospectively. Data regarding individual demographics, presentation, Pap test history, pathology, HPV and p16 positivity, and outcome were analyzed using descriptive statistics.

One pathologist at MD Anderson Cancer Center reviewed tumor pathology. Immunohistochemistry for HPV and p16 was performed using the Bond automated immunostainer. Automated in situ hybridization was performed using the Ventana system.

Back to Top | Article Outline


Six individuals with HPV-positive or p16-positive retroperitoneal unknown primaries or both were identified after review. The median age at diagnosis was 43.5 (range 27-54) (Table 1). Four individuals (66%) presented with lower extremity deep venous thrombosis. All individuals were identified as having pelvic masses on imaging without a known primary. All individuals underwent computed tomography-guided biopsy of their pelvic masses. Four individuals (66%) had no history of abnormal Pap test results. Two individuals (33%) had a remote history of an abnormal Pap test result (more than 10 years), and all individuals had normal Pap test and pelvic examination results at the time of diagnosis. One individual had a hysterectomy for irregular vaginal bleeding performed 3 years before her cancer diagnosis. Her pathology, reviewed at MD Anderson Cancer Center, showed fibroids and no cervical abnormalities.

One individual had liver and lung metastases at the time of diagnosis, whereas the other five individuals had disease confined to the retro-peritoneum. Of note, no individuals had colonoscopies performed as part of their work-up. Two individuals were evaluated by gastroenterology and colonoscopy was not indicated. Other than pelvic imaging with computed tomography or magnetic resonance imaging, these individuals did not have a separate urologic evaluation. No individuals had any head, neck, or thoracic lesions.

Two individuals identified themselves as white (33%), three as Hispanic (50%), and one as African American (17%). Two individuals were nulliparous (33%), whereas the remaining four were parous (66%). One individual had a history of genital warts, and another had a history of Chlamydia. Two individuals were smokers (33%).

Three out of six individuals had squamous features on biopsy (50%), whereas three had undifferentiated (50%) features (Fig. 1). All histology was tested for the presence of HPV or p16 or both (Table 2). Human papillomavirus testing was positive in four (66%) cases, whereas all six (100%) cases were p16-positive (Figs. 2 and 3). Unfortunately, one tissue sample was insufficient to undergo HPV testing.

Two individuals received up-front treatment with taxol and carboplatin chemotherapy. Two individuals received pelvic radiation as initial treatment. One individual received radiation with sensitizing cisplatin. One individual underwent an up-front attempt at resection, which was limited by the location of the mass near pelvic vessels. She then received chemotherapy with carboplatin and radiation. Median follow-up was 12 months (range 6-48 months). Two individuals (33%) have died of their disease, two individuals (33%) have had progression of their disease, and two individuals (33%) are without evidence of disease (Table 3).

Back to Top | Article Outline


Human papillomavirus is a well-known carcinogen in cervical, vulvar, vaginal, anogenital, and oropharyngeal cancers. However, HPV has also been identified in prostate, bladder, breast, esophageal, and lung cancers. In this article, we describe six cases of retroperitoneal masses of unknown primary that were HPV-positive demonstrated either by HPV testing or by p16 testing.

Human papillomavirus is thought to infect cells in these anatomic regions by direct exposure and infection. In malignancies not involving surface or mucosal tissues, the route of transmission in these cases is less clear;1 however, in bladder and prostate cancer, it seems that HPV may be spread directly through sexual transmission.9 In the cases of esophageal and lung cancers, spread may be directly from the oral cavity.4 In breast cancer, HPV has been theorized to spread through the breast ducts.10 HPV has been identified in the sera of a small number of individuals with HPV-positive lung cancer, suggesting that hematogenous spread is possible.11 Hematogenous or lymphatic spread of HPV could explain the findings in the cases we present here. It is also possible that there was microscopic cervical disease that was not detected by current screening or biopsy techniques or that had resolved at primary site.

Human papillomavirus testing has been used to help differentiate between primary gynecologic malignancies when the etiology is unclear. One case report describes how HPV was used to differentiate metastatic cervical cancer to the ovaries from metastatic ovarian cancer.7 In this case, HPV was used along with other tumor markers to determine the etiology. In another series, 89 epithelial ovarian cancers and 29 borderline tumors were negative for HPV.8

Human papillomavirus testing has also been used to differentiate endometrial from endocervical cancers. In a study by Staebler et al,5 24 endometrial and 24 endocervical tumors were examined for HPV positivity via in situ hybridization using a wide-spectrum (HPV 6, 11, 16, 18, 31, 33, 45, and 51) and type-specific probe solution (HPV 16 and 18). Human papillomavirus was positive in 16 out of 24 endocervical cases (66.7%) compared with 0 out of 24 endometrial cases. Furthermore, HPV was then used to help determine the origin of seven uterine adenocarcinomas of unknown origin.5 This same group went on to investigate using p16 immunohistochemistry in 24 endometrial, 19 endocervical, and 4 uterine adenocarcinomas of unknown primary. They found that diffuse p16 expression could distinguish endocervical from endometrial adenocarcinomas and was more sensitive for detecting endocervical carcinomas than HPV in situ hybridization.6 Our case series appears to be the first to our knowledge describing retroperitoneal pelvic masses with unknown primary that are HPV-positive, p16-positive, or both, despite normal cervical examinations and cytology. This is certainly a rare instance, and this may explain why there are no other cases described. Human papillomavirus testing may not be routinely performed in these clinical presentations.

This case series has many limitations. It includes a small number of cases and is retrospective. Also, it is observational in nature. We did not determine the overall rate of HPV positivity in this category of malignancies. In fact, it is not clear how many pelvic masses of unknown primary were tested in our center for HPV. Therefore, the total number of retroperitoneal cancers that were HPV-positive is unknown. We also did not include males in our database search. Also, p16 positivity in tumors outside the lower female genital tract, such as mesothelioma and melanoma, has not been associated with HPV.12 Therefore, p16 is not synonymous with HPV positivity. In our series, however, p16 was positive in all cases in which HPV was positive.

We are not able to conclude that HPV caused these retroperitoneal malignancies in this small case series. It is possible that HPV infected these cells but did not contribute to the development of a malignancy. Lifestyle or genetic factors may have contributed to the development of these malignancies, HPV infection, or both without the two being related. However, HPV testing may be considered for a broader range of malignancies and, if HPV testing is positive, may affect treatment (ie, more likely to perform radiation or chemoradiation).

Given the small number of cases in our study and the various treatment regimens that the individuals received, it is difficult to make a conclusion regarding treatment. In fact, although such individuals might receive a traditional cervical cancer regime for treatment (ie, cisplatin with radiation), four of these individuals received carboplatin and taxol, a more traditional ovarian regimen. Only one individual received cisplatin and radiation and had progression. It does, however, seem reasonable to treat regionally confined cancers such as these with chemoradiation. Practitioners could consider including the uterus and upper vagina in the radiation field to 45 to 50 Gy, because there may be an undetected primary in these regions. Resection followed by chemoradiation may be advisable if possible because it is difficult to control large masses with radiation alone.

This case series of a small number of individuals suggests that HPV may be involved in pelvic masses of unknown primary, even in individuals with normal cervical examination results. Human papillomavirus testing should be considered in the evaluation of such an individual. If HPV testing is positive or negative, then it may help to determine treatment modality.

Back to Top | Article Outline


1. zur Hausen H. Papillomaviruses in the causation of human cancers—a brief historical account. Virology 2009;384:260–5.
2. Dabbs D. Diagnositc immunohistochemistry. 2nd ed. Philadelphia (PA): Churchill Livingstone; 2006.
3. Carozzi F, Confortini M, Dalla Palma P, Del Mistro A, Gillio-Tos A, De Marco L, et al. Use of p16-INK4A overexpression to increase the specificity of human papillomavirus testing: a nested substudy of the NTCC randomised controlled trial. Lancet Oncol 2008;9:937–45.
4. Munoz N, Castellsague X, de Gonzalez AB, Gissmann L. Chapter 1: HPV in the etiology of human cancer. Vaccine 2006;24(Suppl 3):S3/1-10. Epub 2006 Jun 23.
5. Staebler A, Sherman ME, Zaino RJ, Ronnett BM. Hormone receptor immunohistochemistry and human papillomavirus in situ hybridization are useful for distinguishing endocervical and endometrial adenocarcinomas. Am J Surg Pathol 2002;26:998–1006.
6. Ansari-Lari MA, Staebler A, Zaino RJ, Shah KV, Ronnett BM. Distinction of endocervical and endometrial adenocarcinomas: immunohistochemical p16 expression correlated with human papillomavirus (HPV) DNA detection. Am J Surg Pathol 2004;28:160–7.
7. Lee HB, Lee YS, Lee KH, Kim CJ, Park JS. A case of advanced gynecologic pelvic tumors showing the diagnostic utility of HPV analysis. Gynecol Oncol 2009;20:251–3.
8. Idahl A, Lundin E, Elgh F, Jurstrand M, Moller JK, Marklund I, et al. Chlamydia trachomatis, Mycoplasma genitalium, Neisseria gonorrhoeae, human papillomavirus, and polyomavirus are not detectable in human tissue with epithelial ovarian cancer, borderline tumor, or benign conditions. Am J Obstet Gynecol 2010;202:71.e1–6.
9. Boucher NR, Anderson JB. Human papillomavirus and bladder cancer. Int Urogynecol J Pelvic Floor Dysfunct 1997;8:354–7.
10. Kan CY, Iacopetta BJ, Lawson JS, Whitaker NJ. Identification of human papillomavirus DNA gene sequences in human breast cancer. Br J Cancer 2005;93:946–8.
11. Capone RB, Pai SI, Koch WM, Gillison ML, Danish HN, Westra WH, et al. Detection and quantitation of human papillomavirus (HPV) DNA in the sera of patinets with HPV-associated head and neck squamous cell carcinoma. Clin Cancer Res 2000;6:4171–5.
12. Borczuk AC, Taub RN, Hesdorffer M, Hibshoosh H, Chabot JA, Keohan ML, et al. P16 loss and mitotic activity predict poor survival in patients with peritoneal malignant mesothelioma. Clin Cancer Res 2005;11:3303–8.
© 2010 The American College of Obstetricians and Gynecologists