OBJECTIVE: Our purpose was to determine the prognostic significance of human papillomavirus deoxyribonucleic acid in cervical cancers. STUDY DESIGN: The polymerase chain reaction was used to detect human papillomavirus deoxyribonucleic acid types 6, 11, 16, 18, 31, 33, 52, or 58 in tumors from 148 patients (equal numbers of whom were disease free or had relapses) surgically treated for stage IB or IIA cancers in a major Australian hospital. Cox regression modeling was used to assess the effect of human papillomavirus status on tumor recurrence, taking into account patient age, clinical stage, histologic node status, and type of tumor. RESULTS: Seventy of 74 (95%) of the recurring tumors and 62 of 74 (84%) of the nonrecurring tumors were human papillomavirus deoxyribonucleic acid positive. The rates of positivity of types 16 and 18 were 64% versus 31% in the recurrers and 65% versus 14% in the nonrecurrers. Human papillomavirus type 18 positivity was associated with a greater risk of recurrence than was type 16 positivity (hazard ratio 1.8; p = 0.03). Clinical stage, nodal metastasis, and young age (< or = 35 years) also had adverse effects on relapse (hazard ratio for each approximately 2). CONCLUSION: Human papillomavirus type 18 positivity is a risk factor for tumor recurrence in surgically treated cervical cancer.
OBJECTIVE: Our purpose was to determine the prognostic significance of human papillomavirus deoxyribonucleic acid in cervical cancers. STUDY DESIGN: The polymerase chain reaction was used to detect human papillomavirus deoxyribonucleic acid types 6, 11, 16, 18, 31, 33, 52, or 58 in tumors from 148 patients (equal numbers of whom were disease free or had relapses) surgically treated for stage IB or IIA cancers in a major Australian hospital. Cox regression modeling was used to assess the effect of human papillomavirus status on tumor recurrence, taking into account patient age, clinical stage, histologic node status, and type of tumor. RESULTS: Seventy of 74 (95%) of the recurring tumors and 62 of 74 (84%) of the nonrecurring tumors were human papillomavirus deoxyribonucleic acid positive. The rates of positivity of types 16 and 18 were 64% versus 31% in the recurrers and 65% versus 14% in the nonrecurrers. Human papillomavirus type 18 positivity was associated with a greater risk of recurrence than was type 16 positivity (hazard ratio 1.8; p = 0.03). Clinical stage, nodal metastasis, and young age (< or = 35 years) also had adverse effects on relapse (hazard ratio for each approximately 2). CONCLUSION:Human papillomavirus type 18 positivity is a risk factor for tumor recurrence in surgically treated cervical cancer.