Elsevier

Human Pathology

Volume 41, Issue 7, July 2010, Pages 1016-1023
Human Pathology

Original contribution
Human papillomavirus–positive basaloid squamous cell carcinomas of the upper aerodigestive tract: a distinct clinicopathologic and molecular subtype of basaloid squamous cell carcinoma,

https://doi.org/10.1016/j.humpath.2009.11.015Get rights and content

Summary

Basaloid squamous cell carcinoma of the upper aerodigestive tract is a rare, morphologically distinct variant of squamous cell carcinoma that is thought to be clinically aggressive. The histologic features are distinct from, but often confused with, those of human papillomavirus–related oropharyngeal nonkeratinizing squamous cell carcinoma. The role of human papillomavirus as an etiologic agent in true basaloid squamous cell carcinoma is controversial. The purpose of this study was to determine human papillomavirus prevalence and its clinicopathologic significance in upper aerodigestive tract tumors with true basaloid squamous cell carcinoma histology. Twenty-eight cases were identified, 12 in the oropharynx and 16 in the larynx and/or hypopharynx. High-risk human papillomavirus in situ hybridization and immunohistochemistry for p16 and p53 were performed. Nine (75%) of the oropharyngeal and none of the larynx/hypopharynx tumors were human papillomavirus positive. Human papillomavirus–positive tumors affected younger patients. No significant statistical differences in patients' sex, tumor stage, treatment modality, or length of follow-up were observed between the 2 groups. Viral status showed a strong, positive correlation with p16 (P < .001) and a strong, negative correlation with p53 (P < .0001) immunoreactivity. Overall survival was better for human papillomavirus–positive basaloid squamous cell carcinomas (P < .05), with 86% of patients alive at 3 years compared with 35.3% of patients with human papillomavirus–negative tumors. These findings suggest that a subset of basaloid squamous cell carcinomas is virally driven. These tumors occur almost exclusively in the oropharynx, are molecularly distinct from their human papillomavirus–negative counterparts, and have a more favorable clinical outcome.

Introduction

Basaloid squamous cell carcinoma (BSCC) of the upper aerodigestive tract (UADT) was originally described by Wain et al [1] in 1986 as a rare variant of squamous cell carcinoma (SCC) characterized by clinically aggressive behavior and distinct histologic features. These tumors have a predilection for the larynx, the hypopharynx, and, less frequently, the oropharynx [2]. Rarely, it may occur at other head and neck sites [3], [4]. Like conventional SCC, BSCC is strongly associated with traditional risk factors such as tobacco and alcohol abuse [2], [5]. The prognosis of BSCC is reported by some to be worse than that of conventional SCC independent of tumor stage, although others have found no difference stage for stage [6], [7], [8], [9], [10], [11], [12].

Two major histopathologic features define BSCC. The first is a basaloid component consisting of a solid, lobular growth of small, crowded cells with scant cytoplasm, hyperchromatic round nuclei, and small cystic spaces containing PAS- or alcian blue–positive myxoid material. The second feature is an intimate association with SCC or dysplasia or the presence of focal squamous differentiation. Other accessory findings, including small or large central foci of coagulative necrosis and stromal hyalinosis, are also described [1] (Fig. 1).

These histologic features are distinct from, but may be confused with, those of oropharyngeal human papillomavirus (HPV)–related nonkeratinizing squamous cell carcinoma (NK SCC) [13], [14]. In contrast to BSCC, HPV-related NK SCC occurs predominantly in the tonsillar tissue of the oropharynx and much less frequently at other head and neck sites. It tends to occur in younger patients and have unique risk factors related to certain sexual behavior, including multiple partners, early age at first intercourse, and oral sex. HPV-related NK SCC also has a more favorable prognosis [15], [16], [17], [18], [19], [20], [21]. The tumor cells in NK SCC are ovoid to spindle-shaped with hyperchromatic nuclei that lack prominent nucleoli and have indistinct cell borders (Fig. 2). NK SCC is also characterized by a lack of prominent stromal response. Stromal hyalinosis and cystic spaces containing mucin-like material, often found in BSCC, are not seen in NK SCC. The cells in BSCC, unlike NK SCC, adopt an organized “jigsaw puzzle” growth pattern. In addition, association with surface squamous dysplasia beyond the tonsillar crypt lining is unusual in NK SCC.

Whether or not HPV plays a role in the pathogenesis of BSCC of the UADT is controversial. Using polymerase chain reaction (PCR) amplification and in situ hybridization (ISH) techniques, Cabanillis et al [22] found no evidence of HPV DNA in 9 cases of BSCC of the pharynx, 5 of which were oropharyngeal and 4 hypopharyngeal. In another study, no HPV DNA was identified by ISH in 9 cases of BSCC, 7 from the larynx/hypopharynx and 2 from the base of tongue [23]. In contrast, Kleist et al (2004) [24] found 9 of 10 BSCCs of the oropharynx, larynx, and hypopharynx to be HPV positive by PCR. More recently, in a study of BSCC of the UADT, Begun and Westra [25] identified HPV 16 by ISH in 16 (76%) of 21 cases of BSCC of the oropharynx but only in 2 (6%) of 32 from nonoropharyngeal sites. Such discrepancies between different studies may have resulted from lack of use of strict histologic criteria for identifying the true BSCC. Furthermore, in tumors that were positive for HPV, it was not clear whether the virus was involved in the pathogenetic process or was merely a coincidental finding.

A distinct molecular profile distinguishes SCCs that are HPV driven from HPV-unrelated tumors. This profile may be useful in identifying tumors in which HPV is an etiologic agent from tumors in which HPV is merely a bystander. HPV-driven tumors overexpress the tumor suppressor protein p16 and commonly lack p53 mutations [26], [27], [28]. p16 is a cell cycle protein that prevents the inactivation of retinoblastoma protein and progression of the S-phase of the cell cycle. It is thought that HPV oncoprotein E7–mediated inactivation of retinoblastoma protein leads to a loss of inhibition of p16 transcription and its subsequent overexpression. p16 overexpression is considered a surrogate marker of HPV-induced tumors of the cervix and UADT and is distinctly uncommon in HPV-negative SCC [13], [14], [27], [28], [29]. HPV-negative SCC also frequently harbors p53 mutations leading to accumulation of p53 protein that is detectable by immunohistochemistry [13], [14], [28]. On the other hand, in most HPV-induced tumors, wild-type p53 protein is targeted for ubiquitination and degradation by the HPV E6 oncoprotein leading to little detectable protein by immunohistochemistry [13], [28]. Thus, strong and diffuse immunostaining for p16 combined with focal or negative p53 immunostaining is considered to be a molecular signature for HPV-induced SCC.

The purpose of this study was 3-fold: first, to determine the prevalence of high-risk HPV in oropharyngeal and nonoropharyngeal UADT tumors with true BSCC histology as described by Wain et al [1]; second, to explore whether HPV is a causative agent in HPV-positive tumors as determined by the expression pattern of the key cell cycle proteins, p16 and p53; and third, to investigate the clinical significance of HPV in BSCC in terms of patient outcomes.

Section snippets

Case selection

The study was approved by the Human Studies Committee of Washington University. The surgical pathology files at Barnes-Jewish Hospital, Washington University, were searched for patients diagnosed with BSCC of the head and neck between 1988 and 2007. Hematoxylin- and eosin-stained slides from primary tumors that had corresponding paraffin-embedded tissue were retrieved and reviewed by 2 of the study pathologists (R. D. C. and S. K. E. M.). The diagnosis of BSCC was based on the criteria

Results

Twenty-eight cases of BSCC, 12 from the oropharynx and 16 from the larynx/hypopharynx, were identified. The age of the patients ranged from 40 to 68 years, and the tumors were more common in men (64.3%) than women (35.7%). Most tumors were stage III or IV, with only 3 stage I tumors and 1 stage II tumor. Staging information was not available for 1 patient. Treatment modality was variable, although surgery followed by postoperative radiation was most common. The median length of follow-up was 26

Discussion

Using strict microscopic criteria for diagnosis of BSCC and excluding all cases with ambiguous features, we found an HPV prevalence of 75% in BSCC of the oropharynx. At the same time, none of the laryngeal/hypopharyngeal tumors was positive. These results parallel those of Begum and Westra [25] who found 76% of oropharyngeal and 6% of nonoropharyngeal BSCCs to be HPV positive by ISH.

The prevalence of HPV in BSCC of the UADT varies widely in different studies, with some authors failing to detect

Acknowledgments

We thank Kevin Keith for his expert technical assistance with immunohistochemistry and ISH experiments.

References (32)

  • ErdamarB. et al.

    Basaloid squamous cell carcinoma of the supraglottic larynx

    Eur Arch Otorhinolaryngol

    (2000)
  • de Sampaio GoesF.C. et al.

    Prognoses of oral basaloid squamous cell carcinoma and squamous cell carcinoma: a comparison

    Arch Otolaryngol Head Neck Surg

    (2004)
  • LunaM.A. et al.

    Basaloid squamous carcinoma of the upper aerodigestive tract. Clinicopathologic and DNA flow cytometric analysis

    Cancer

    (1990)
  • ThariatJ. et al.

    Outcomes after radiotherapy for basaloid squamous cell carcinoma of the head and neck: a case-control study

    Cancer

    (2008)
  • WienekeJ.A. et al.

    Basaloid squamous cell carcinoma of the sinonasal tract

    Cancer

    (1999)
  • El-MoftyS.K. et al.

    Prevalence of human papillomavirus type 16 DNA in squamous cell carcinoma of the palatine tonsil, and not the oral cavity, in young patients: a distinct clinicopathologic and molecular disease entity

    Am J Surg Pathol

    (2003)
  • Cited by (0)

    The authors also wish to acknowledge the support of the Biostatistics Core, Siteman Comprehensive Cancer Center (St. Louis, MO, USA), and NCI Cancer Center Support Grant P30 CA091842.

    Presented at the United States and Canadian Academy of Pathology, 98th annual meeting, Boston, MA, 2009.

    View full text