The aim of the study was to evaluate the immunohistochemical expression of p16INK4a as a marker of progression risk in low-grade dysplastic lesions of the cervix uteri. p16INK4a immunohistochemistry was performed on 32 CIN1 with proven spontaneous regression of the lesion in the follow-up (group A), 31 (group B) with progression to CIN3 and 33 (group C) that were randomly chosen irrespective of the natural history of the lesion. p16INK4a staining pattern was scored as negative (less than 5% cells in the lower third of dysplastic epithelium stained), as focally positive (< or = 25%) and as diffuse positive (> 25%). A diffuse staining pattern was detected in 43.8% of CIN1 of group A, 74.2% of group B and 56.3% of group C. No p16INK4a staining was detected in 31.3% and 12.9% CIN1 lesions of groups A and B, respectively. Overall, 71.4% and 37.8% of p16INK4a-negative and diffusely positive CIN1 had regressed at follow-up, whereas 28.6% and 62.2% negative and diffusely positive CIN1 were progressed to CIN3, respectively (P < 0.05). All CIN3 lesions analyzed during follow-up of group B were diffusely stained for p16INK4a. Although p16INK4a may be expressed in low-grade squamous lesions that undergo spontaneous regression, in this study, CIN1 cases with diffuse p16INK4a staining had a significantly higher tendency to progress to a high-grade lesion than p16INK4a-negative cases. p16INK4a may have the potential to support the interpretation of low-grade dysplastic lesions of the cervix uteri.