A series of 532 women with genital HPV infections has been prospectively followed-up without treatment since 1981 for a mean of 50 (+/-21) months. The patients were examined at six month intervals by colposcopy, PAP smears and/or biopsy. HPV typing of all biopsies was completed using in situ, Southern blot and/or sandwich hybridization with DNA probes for types 6, 11, 16, 18, 31 and 33. Survival data analysis was applied to analyse the clinical course (i.e. spontaneous regression and progression) of the HPV lesions stratified by their HPV type, currently available for 458 women. Clinical progression was significantly related to the HPV type present in the lesions. The progression rate was 11.1% (6/54) for HPV 6 lesions, 14.3% (8/56) for HPV 11, 35.2% (32/91) for HPV 16, 12.5% (4/32) for HPV 18, 18.8% (6/32) for HPV 31, 19.4% (6/31) for HPV 33 and 28.6% (4/14) for doubly infected lesions. The lowest progression rate, 6.1% (9/148), was found in lesions which remained constantly HPV DNA-negative. In the survival analysis the probability of progression varied significantly between the six HPV types (p = 0.0005, overall). After grouping the viral types as HPV 6/11 ('low risk'), HPV 16/18 ('high risk') and HPV 31/33 ('intermediate risk') the overall probability of progression remained significantly different (p = 0.0035, overall). In clinical regression, however, the HPV type was not an equally good predictor (p = 0.1952, overall). Within groups HPV 6/11, 16/18 and 31/33 the differences were even less significant (p = 0.4759, overall). In the pairwise comparison significant differences in progression occurred when HPV type 16 was compared to HPV 6, HPV 11 or HPV DNA-negative lesions.(ABSTRACT TRUNCATED AT 250 WORDS)