French study shows that HAART makes little difference to women's risk of developing pre-cancerous cervical lesions

HIV-positive women remain at high risk of developing pre-cancerous lesions in the cervix, even if they are taking potent antiretroviral therapy, according to a French study published in the December edition of Antiviral Therapy. The investigators write that their findings underline the need for HIV-positive women “to participate in cervical cancer screening programmes with at least an annual visit whether or not they are on HAART.”

Thanks to potent anti-HIV therapy, there has been a massive decline in the incidence of most HIV-related opportunistic infections in richer countries. The impact of antiretroviral therapy on anal and cervical squamous intraepithelial lesions in HIV-positive individuals is controversial.

Pre-cancerous lesions in the anus and cervix appear to occur more frequently in HIV-positive individuals. Pre-cancerous and cancerous cell changes in the anus and cervix are almost always associated with human papilloma virus (HPV) infection. Although some studies have found a link between a low CD4 cell count and a higher risk of pre-cancerous lesions, others have not.

Investigators in Paris designed a prospective study involving women receiving care at two specialist HIV clinics. They measured the incidence of pre-cancerous cervical lesions. They then compared the incidence of these lesions in women treated with potent HIV therapy to that observed in women who either received their care before effective anti-HIV therapy became available, or in women who were not taking antiretroviral treatment. The impact of CD4 cell count and behavioural risk factors on the incidence of pre-cancerous cervical lesions was then assessed.

Although 1,124 women expressed an interest in participating in the study, most were ineligible for participation, mainly because of pre-existing cervical abnormalities. Nearly 50% of women were excluded because they had a previous or current pre-cancerous cervical lesions. A further 46 were excluded because colposcopy identified abnormalities at baseline, and 173 were excluded because of a lack of follow-up data.

This left the investigators with a study population of 298. Every six months these women had pelvic examinations, Pap smears, colposcopic examinations, and answered a structured questionnaire. Information was also obtained on the patients’ demographics, use of antiretroviral therapy, and their CD4 cell count and viral load closest to each visit.

Demographic and baseline characteristics of the study population changed markedly during the course of the study. Women enrolled after 1997 were significantly more likely to be older (34 vs. 32 years, p = 0.01), from sub-Saharan Africa (54% vs. 31%, p = 0.005), and to report inconsistent or no condom use (82% vs. 68%, p = 0.01), but they were less likely to smoke (22% vs. 47%, p

A CD4 cell count of 200 cells/mm³ or less, the point at which an individual becomes vulnerable to some life-threatening AIDS-defining illnesses, was present in women at 14% of study visits. The investigators note “this low proportion suggests that only few among the patients included in the current study were severely immunocompromised.”

Amongst the women enrolled after 1997, the median CD4 cell count was 400 cells/mm³, and the median viral load was 500 copies/ml. Two-thirds of women entering the study after 1997 were treated with potent anti-HIV therapy, and of 113 women who had at least two visits after commencing HIV therapy, median viral load fell from 199 copies/ml at the first visit to below 50 copies/ml at the second. At the same time, CD4 cell count increased from a median of 301 cells/mm³ to 434 cells/mm³. These changes were statistically significant (p = 0.001 and p = 0.0006, respectively.

The cumulative incidence of pre-cancerous cervical lesions was 17%. Patients were followed for a median of 28 months, providing a total of 1,013 person-years of follow-up. In this period 88 women developed pre-cancerous cervical lesions, providing an incidence rate of 8.7 per 100 person years.

At the time pre-cancerous lesions were detected, median CD4 cell count was 417 cells/mm³ and median viral load was 499 copies/ml. A total of 31 (35%) of women developing pre-cancerous cervical lesions were taking potent antiretroviral therapy.

In most cases (75), the lesions were classed as low-grade. Eight of the 13 women with incident high-grade lesions had had a normal Pap smear in the ten months before diagnosis. No cases of invasive cervical cancer were diagnosed.

Amongst women receiving potent HIV therapy, the incidence of pre-cancerous cervical lesions was 6.5 cases per 100 person years, compared to 10.7 cases per 100 person years in the women not receiving powerful combinations of HIV treatment. In univariate analysis this difference was statistically significant (p = 0.02). Univariate analysis also identified age under 40 at enrolment (p = 0.005), tobacco smoking (p = 0.008), more than five lifetime sexual partners (p = 0.01), use of the contraceptive pill (p = 0.003), inconsistent or no condom use (p = 0.02), enrolment before 1997 (p = 0.02) and not taking potent anti-HIV therapy (p = 0.04) as being associated with the development of pre-cancerous cervical lesions.

However, when the investigators performed multivariate analysis, only age under 40 remained a significant predictor for the development of pre-cancerous cervical lesions (p = 0.02). Although the risk of cervical lesions was slightly reduced in women taking potent HIV therapy compared to those who were not (risk ratio, 0.7), this reduction did not meet statistical significance (p = 0.15).

“This study indicates that HIV-positive women have a high risk of developing squamous intraepithelial lesions and suggests little if any benefit from HAART on the occurrence of cervical lesions”, write the investigators.

They add, “our findings have implications for the gynaecological care of HIV-positive women.”