A comparison of slot blot, southern blot, and in situ hybridization analyses for human papillomavirus DNA in genital tract lesions.

Genital tract lesions were analyzed for human papillomavirus (HPV) DNA by in situ hybridization using probes of HPVs 6/11, 16/18, and 31/33/35. All of the HPVs detected in vulvar and perianal condylomata by in situ hybridization were HPV 6/11-related, whereas the majority of HPVs detected in cervical intraepithelial grade I lesions were types 16/18- and 31/33/35-related. None of the lesions with histologic features equivocal for HPV infection had detectable HPV DNA by in situ hybridization, though some did contain HPV DNA sequences as ascertained by filter hybridization analysis. The sensitivity of in situ hybridization was compared with that of filter hybridization (slot blot and/or Southern blot). The correlation was high (28 of 30) for cases that contained HPVs 6/11 or 16, as deduced by filter hybridization, and was much less (ten of 29) for cases that contained HPVs distinct from types in the filter hybridization probe cocktail (HPVs 6/11, 16, 18, 31, 35 and 51). There was a high concordance between the results of Southern blot hybridization and slot blot hybridization analyses, especially with cases that contained HPVs 6/11 and 16. In situ hybridization, slot blot, and Southern blot hybridization analyses are all very effective in detecting the common HPV types (HPVs 6/11 or 16). In situ hybridization is useful in differentiating cervical lesions that contain HPV 6/11 from those that contain HPV 16 or other types with oncogenic potential. However, filter hybridization is superior to in situ hybridization when analyzing cases with histologic findings equivocal for HPV infection or cases that contain HPV types related to, but distinct from, the types included in the probe.