Performance of the Roche cobas 4800 high-risk human papillomavirus test in cytologic preparations of squamous cell carcinoma of the head and neck

Authors

  • Darcy A. Kerr MD,

    1. Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts
    2. Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • Martha B. Pitman MD,

    1. Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts
    2. Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • Brenda Sweeney MS, SCT(ASCP) MBCM,

    1. Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts
    2. Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • Ronald N. Arpin III MS, SCT(ASPC),

    1. Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts
    2. Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • David C. Wilbur MD,

    1. Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts
    2. Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • William C. Faquin MD, PhD

    Corresponding author
    1. Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts
    2. Harvard Medical School, Boston, Massachusetts
    • Corresponding author: William C. Faquin, MD, PhD, Department of Pathology, Warren 219, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114; Fax: (617) 573-3389; wfaquin@partners.org

    Search for more papers by this author

  • The authors thank Dr. J. Rocco for his valued assistance.

Abstract

BACKGROUND

Determining high-risk human papillomavirus (HR-HPV) status of head and neck squamous cell carcinoma (HNSCC) defines a tumor subset with important clinical implications. Cytologic sampling often provides the sentinel or sole diagnostic specimen. The authors assessed the performance characteristics for the Roche cobas 4800 HPV real-time polymerase chain reaction (PCR)-based system (cobas) on cytologic specimens of HNSCC compared with standard methods of in situ hybridization (ISH) for HR-HPV and immunohistochemistry (IHC) for p16 on formalin-fixed, paraffin-embedded (FFPE) tissue.

METHODS

Samples of HNSCC were collected by fine-needle aspiration and from surgical biopsies or resections, fixed, and processed with the cobas system. Available corresponding FFPE samples were synchronously evaluated for HR-HPV using ISH and IHC. Discrepant cases underwent additional PCR studies for adjudication.

RESULTS

Thirty-six samples from 33 patients were analyzed. Forty-two percent (n = 15) of tumors were positive for HR-HPV according to cobas. Corresponding histology with ISH (n = 30) was concordant in 91% of samples. Compared with the adjudication PCR standard, there were 3 false-positive cases according to cobas. Ninety-two percent (n = 12) of cases were the HPV16 subtype. The overall sensitivity for the cobas system was 100%, and the specificity was 86%.

CONCLUSIONS

Concordance in HNSCC HR-HPV status between cobas and ISH/IHC was > 90%, and cobas demonstrated a sensitivity of 100% and a specificity of 86%, broadening options for HR-HPV testing of fine-needle aspiration samples. Advantages for this system include subtyping of HR-HPV and the ability to discern HR-HPV status earlier in a patient's treatment course. Cancer (Cancer Cytopathol) 2014;122:167–174. © 2013 American Cancer Society.

Ancillary