High‐risk HPV positivity is a long‐term risk factor for recurrence after cervical excision procedure in women living with HIV

Abstract Objective To evaluate the risk factors for recurrence of high‐grade disease after cervical excision in women living with HIV (WLWH), with a specific interest in the role of high‐risk (HR‐) HPV positivity. Methods Multicentric retrospective study conducted on WLWH who underwent cervical excision between January 1987 and June 2017 in six Italian institutions. The rate of high‐grade recurrence was determined. Risk factors for recurrence and HR‐HPV positivity were determined with the Log‐rank test and Cox proportional hazards regression models. Results A total of 271 WLWH were included in the final analysis. A high‐grade recurrence was found in 58 (21.4%) patients. Age 41 years or more at inclusion and HR‐HPV positivity during follow up were independently associated with a higher risk of disease recurrence with relative risks of 4.15 (95% confidence interval [CI] 2.01–8.58, P < 0.001) and 5.18 (95% CI 2.12–12.67, P < 0.01), respectively. Age 41 years or more (relative risk 1.75, 95% CI 1.01–3.04, P = 0.047) resulted as a risk factor for HR‐HPV positivity during follow up. Conclusion HR‐HPV positivity is a risk factor for recurrence after cervical excision in WLWH. Women older than 41 years may benefit from a long‐term yearly follow up. Future studies regarding HPV vaccination after treatment in WLWH may be useful, considering the protective role of the higher probability of HPV negativity in vaccinated women.


| INTRODUC TI ON
Human papillomavirus (HPV) infection is the most common sexually transmitted infection, with over 300 million infected women globally. Risk factors for acquiring HPV infection include multiple sexual partners, immunodeficiency status, early sexual debut, parity, use of contraceptives, and smoking. 1 Compared with the general female population, women living with HIV (WLWH) have a higher prevalence of high-risk HPV (HR-HPV) infection, HR-HPV persistent infection, higher risk of cervical intraepithelial neoplasia (CIN) lesions, and higher incidence of cervical cancer. 2,3 The presence of HIV and related immunosuppression significantly affects the natural history of HR-HPV infection, increasing the susceptibility to acquiring HR-HPV, reducing the rate of clearance, and favoring both the reactivation of latent infection and the persistence of HR-HPV infection. 2,3 Severe immunosuppression, as reflected by low CD4 + T-cell count or increased HIV loads, has been consistently associated with the risk of HR-HPV infection itself and with the risk of preinvasive and invasive cervical lesions. 3 Highly active antiretroviral therapy (HAART), together with the associated immune reconstitution, seems to improve the control of HR-HPV infection and disease progression. 3 To date, because primary prevention with HPV vaccination in WLWH, although promising, still does not have strong scientific validation and further studies are undoubtedly necessary, 4,5 cervical cancer prevention in WLWH is mainly based on secondary prevention, with early diagnosis and treatment of high-grade CIN (CIN 2/3).
Surgical cervical excision (i.e. loop electrosurgical excision procedure, laser conization, or cold-knife conization) is the recommended procedure in the case of CIN 2/3. 6 Although the surgical treatment of cervical cancer precursors is highly effective in immunocompetent women, with eradication in about 90% of cases, a lower efficacy has been reported in WLWH, in whom high rates of persistent and recurrent disease have been found in several studies. 7 For this reason, a long-term follow up is necessary for WLWH subjected to cervical excision for intraepithelial neoplasia. 8 However, there is currently no consensus as to what the follow-up method after cervical excision in WLWH should be. Available guidelines contain recommendations ranging from the same methods of follow up as HIV-negative women to annual follow up, particularly in the case of severe immunodeficiency. 6,9,10 Evidence suggests that several factors may be associated with increased rates of persistent and recurrent disease after treatment, including smoking, positivity of cone margins, persistent HR-HPV infection, and immunosuppression. 11 Persistent HR-HPV infection is gaining increasing importance in the evaluation of the recurrence risk after treatment in HIV-negative women, 12 but evidence in this regard is scarce for WLWH. Knowing the impact of HPV infection on the risk of recurrence after treatment might also be the basis for future studies on the role of HPV vaccination after treatment in WLWH.
The aim of the present study was, therefore, to evaluate the risk factors for recurrence of high-grade lesions after cervical excision in WLWH, with a specific interest in the role of HR-HPV positivity during follow-up.

| MATERIAL S AND ME THODS
This was a multicentric, retrospective, cohort study that involved six Italian institutions. All WLWH who underwent cervical excision between January 1987 and June 2017 in the six institutions were retrospectively identified by searching the clinical databases and included in the present study.
The cervical treatment was performed because of high-grade squamous intraepithelial lesions (HSIL, CIN2/3) or persistent (>2 years) low-grade squamous intraepithelial lesions (LSIL, CIN1) diagnosis. Only women with a definitive diagnosis of high-grade intraepithelial lesion (HSIL, CIN2/3) at the cone specimen were included in the final analysis. Women who reported previous treatment for cervical pathology were excluded. Likewise, patients with missing data and with a diagnosis of cancer on the first visit or after conization were not included in the analysis.
We collected pertinent sociodemographic and clinical data such as age, menopausal status, age at first sexual intercourse (≤16 years or ≥17 years), number of sexual partners (up to five or six or more), tobacco use, parity (nulliparous or parous), and reported route of transmission of HIV infection. Moreover, we collected data regarding the referral cytology, the colposcopic examination at inclusion, the histopathologic findings on pre-operative biopsy, the final histopathologic diagnosis at the cone specimen, and the HR-HPV status before cervical treatment and during follow up.
Abnormalities on referral cervical cytology were classified according to the most recent Bethesda system terminology. 13 Cytology examinations performed before the introduction of the most re-  The log-rank test of risk factors for HR-HPV positivity during follow up is reported in Table 2. Age between 30 and 35 years and a WHO Class "Not significant" at follow up were associated with a lower risk of HR-HPV positivity, whereas age of 41 years or older, smoking, a "Severe" WHO Class at inclusion, and a WHO Class "Severe" at follow up were associated with a higher risk of HR-HPV positivity during follow up. All these factors were included in a Cox proportional-hazards regression model. Only one factor related to a higher risk of HR-HPV positivity during follow up was retained in the model: age 41 years or more, with an RR of 1.75 (95% CI 1.01-3.04, P = 0.047).

| DISCUSS ION
The results from the present study showed that cervical excision is an effective procedure for high-grade CIN treatment in WLWH, considering that 78.6% of women presented a long-term negative follow up. However, the risk of recurrent high-grade lesion or invasive cancer seems to last for many years. In our cohort, women aged 41 years or more had a four times higher risk of recurrence of HSIL+, and women with HR-HPV positivity during follow up had a five times higher risk. Moreover, age 41 years or more was found to be a risk factor for HR-HPV positivity during follow up. High-risk HPV positivity during follow up is a recognized risk factor for both short-term and long-term recurrence of cervical lesions after treatment in HIV-negative women. 11 This association has been evaluated less frequently in WLWH, and data are not as consistent.
Indeed, whereas Massad et al. 8 reported a hazard ratio of 2.9 in case of HPV positivity for recurrence of any grade, Lodi et al. 18 affirmed that high-risk HPV subtypes were detected in most cases but were not associated with recurrence.
Immunosuppression, expressed as low CD4 + T-cell count, is considered one of the most significant risk factors for disease recurrence in WLWH. 3,19 In our study, CD4 + T-cell count at the time of cervical treatment or at the time of the last control was not associated with a higher risk of recurrence. Literature data about the most appropriate prognostic CD4 + T-cell count are conflicting. Clifford et al. 19 showed that nadir CD4 + T-cell count was a more discriminant measure of risk for CIN2+ than CD4 + T-cell count at diagnosis. Other studies found that CD4 + T-cell count at the time of treatment is a better predictor of persistent disease or recurrence than nadir CD4 + T-cell count. 8 Our results could be interpreted as the fact that the pathogenesis of highgrade cervical lesions is multifactorial, and immunodeficiency could play a role as a cofactor, but not as a determining factor. Previous In the WLWH included in our cohort, HR-HPV test during follow up after cervical excisional treatment presented a high negative TA B L E 1 Logrank test for factors associated with cumulative disease recurrence at 180 months in 271 women with diagnosis of CIN2/3 at the cone specimen a Abbreviations: ART, antiretroviral therapy; CIN, cervical intraepithelial neoplasia; HAART, highly active antiretroviral therapy; HR, high-risk; LEEP, loop electrical excision procedure.
a Values are given as number (percentage) and hazard ratio (95% confidence interval). b High-risk-HPV was performed in 126 women at inclusion. women. 23 If this approach proves effective, it could be evaluated with further studies also in WLWH.
The increased risk of high-grade recurrence for patients aged 41 years or older that emerged from our data could be explained by the age-related immunosenescence of older WLWH that is reported to occur earlier than in HIV-negative women of the same age. 24 Moreover, age has been reported as a risk factor for recurrence after cervical excision also in HIV-negative women, although it may be of less importance than other factors. 25 In our study, we did not find any statistically significant association with sociodemographic characteristics (except age), procedural modalities, or margin involvement. It is interesting to note that smoking was not a risk factor for disease recurrence in our cohort, unlike what is reported in the literature. 21,22 This lack of association could be related to the high percentage of smokers in our population (68.3%).
As also reported in a recent meta-analysis, 11 HPV positivity during follow up seems to play a more important role than the positive margins in predicting disease recurrence. This may be even more relevant in a cohort of immunocompromised patients like WLHW, such as those included in our study.
The major strengths of our study are the large cohort of WLWH However, the presented data about long-term follow up are necessary to deepen the history of these patients. Moreover, it also needs to be acknowledged that CD4 + count at the time of treatment may not have the same clinical significance for each included WLWH, and generalization from these data may be limited from the heterogeneous immunologic management of the included women because of temporal changes in HAART use.
In conclusion, surgical excision of CIN seems to be an effective procedure in WLWH, with a high rate of negative follow up.
However, a careful, intensive, and prolonged follow up after treatment is necessary. As the main risk factor for long-term recurrence seems to be HR-HPV positivity during follow up, the implementation of this test is needed in WLWH with a history of cervical excisional treatment with more systematic use of co-testing.

F I G U R E 2
Kaplan-Meier curve for disease recurrence according to age 41 years or more or less than 41 years at the time of treatment F I G U R E 3 Kaplan-Meier curve for disease recurrence according to high-risk-HPV positivity during follow up In our opinion, the interval of follow up after treatment could be individualized according to the patient's age. Whereas in young women it is possible to adopt a follow-up interval similar to that used in HIV-negative women, older women should be followed at yearly intervals after 24 months from treatment.

CO N FLI C T S O F I NTE R E S T
The authors have no conflicts of interest.  aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.