HPV oral lesions in HIV-infected patients: the impact of long-term HAART

Introduction

A hallmark of the HIV-associated oral lesion spectrum is the increase in the frequency of human papillomavirus-associated oral lesions (HPV-OL), concurring with the introduction of highly active antiretroviral therapy (HAART) 1-3. The term HPV-OL incorporates a group of benign oral lesions (squamous cell papilloma, verruca vulgaris, condyloma acuminatum, and multifocal epithelial hyperplasia) that has been described in HIV-infected patients as a result of diverse immunological abnormalities 4-6.

Human papillomavirus-associated oral lesions in immunocompromised patients may carry unusual HPV-types and atypical clinical features 7. In most studies 8-14, low-risk HPV types have been associated with these oral benign disorders; however, some high-risk types (HPV-16, 18, 31) have also been identified 8, 9, 12, 15.

Although the benign nature of HPV-OL in HIV has been documented 16-18, it is relevant to consider the increased incidence of diverse types of cancer with a known infectious cause reported in HIV/AIDS patients, particularly HPV-associated oral and oropharyngeal cancer 4, 19-25. Thus, the aim of this study is to determine the prevalence, clinical characteristics, and factors associated with the presence of HPV-OLs in Mexican HIV-infected patients, as well as its genotyping in the HAART era.

Material and methods

A cross-sectional study was conducted from July 2004 to December 2011 in a primary care center for HIV/AIDS patients without Social Security, located in Mexico City (Clínica Especializada Condesa [CEC]). The study protocol was approved by the institutional review board of the participant institution, and the patients gave written consent to participate in the study.

Adult patients with known HIV infection, who attended a routine visit at the Oral Medicine and Pathology Clinic of CEC were consecutively included. Information regarding age, HIV-transmission risk factors, and tobacco use was recorded. Lymphocyte CD4⁺ T-cell counts, HIV viral load, CDC stage 26, and the type and duration of HAART were retrieved from the medical records. HAART was defined as the concomitant use of three antiretroviral (ARV) drugs, either a combination of two types of ARVs (nucleoside reverse transcriptase inhibitors [NRTIs], non-nucleoside reverse transcriptase inhibitors [NNRTIs], protease inhibitors [PIs], or a fusion inhibitor [FIs]), or three NRTIs 27. Patients were considered to be on HAART if they received any of these combinations for more than 30 days; the regimens were supervised by an infectious disease specialist.

All the patients had a thorough clinical oral examination; HPV-OLs were identified by two oral pathology and medicine specialists with experience in the recognition of these lesions (VRA, GAS), based on visual inspection and according to predetermined clinical criteria 7, 28, 29. The presence and clinical characteristics of each HPV-OL were registered. Tissue from the HPV-OL was obtained using a disposable punch, and divided into two sections: one was fixed in a formalin solution (10%) for subsequent routine histopathological analysis (hematoxylin–eosine stain); the second piece was introduced in a collection tube that contained a cellular preservation solution (CytycCorp., Marlborough, MA, USA).

Histopathological diagnosis was based on pre-established definitions 7, 29 that consider the morphology of the epithelial surface, the degree of keratinization, basal cell hyperplasia, mitosoidal body presence, vacuolization of squamous cells, and inflammation in the subjacent connective tissue. All studies were performed by oral pathology and medicine specialists blinded to HPV testing.

Specimens were vortexed to remove the cells from the bristles and then they were centrifuged at 3000 g for 10 min. The cell pellets were resuspended in 200 μl of sterile PBS. Both cellular and tissue specimens were immediately frozen at −20°C until DNA extraction was performed. DNA was extracted using the DNA Wizard Genomic Purification kit (Promega, Inc., Madison, WI, USA) according to the manufacturer's protocol. The proteins were removed by incubation at 65°C in a nucleic lysis solution for 3 h (brushings) or overnight (biopsies) with 17.5 μl of proteinase K (20 mg/ml); then, 1.5 μl of RNase was added and incubated at 37°C for 30 min. One hundred microliters of protein precipitation solution was added, vortexed for 20 s, and kept in ice for 5 min. A white pellet was obtained after 4 min of centrifugation, washed with 300 μl of isopropanol, and then with 300 μl of 70% ethanol. Following ethanol precipitation, the DNA was resuspended in 50 μl of DNA rehydration solution. Finally, the obtained DNA was stored at −20°C in 5-μl aliquots until analysis.

The total DNA concentration was determined by spectrophotometry, the ratio of 260/280 was measured to evaluate the DNA purity using a Beckman DU-640 spectrophotometer (Beckman Scientific Instruments, Fullerton, CA, USA). To demonstrate the integrity of the purified DNA, a β-globin gene PCR assay (268-pb PCO4/6H20) was performed for all samples.

Genomic HPV-DNA amplification was carried out using MY09/MY11 30 and GP5+/GP6+ primers 31 that amplify a large variety of HPV types, under the conditions described by the manufacturer, as previously described 32. In every assay, water and DNA from the HeLa cell line was used as a substitute for DNA and as a negative and positive control, respectively. To visualize the PCR products after amplification, a 5-μl aliquot of each sample was run on a 1.5% agarose gel staining with ethidium bromide (10 mg/ml) (Sigma Chemical Co., St. Louis, MO, USA); the expected sizes were ~450 bp for MY09/11 and ~150 bp for GP5+/6+.

To minimize sample-to-sample contamination, special precautions such as frequent changes of gloves and careful handling of samples were followed. All the reagents used in the PCR assays were prepared and stored in an area that was free of PCR-amplified products.

Positive PCR products were purified using the QIAquick PCR purification protocol and subsequently sequenced using the Big Dye Terminator v3.1 cycle sequencing kit (Applied Biosystems Life Technology, Foster City, CA, USA) by using one of the PCR primers as a sequencing primer (GP5+ and MY09). Sequencing reactions were analyzed using the ABI PRISM 3100 genetic analyzer system (Applied Biosystems Life Technology), and the obtained sequences were matched with Gene Bank-published sequences (National Center for Biotechnology Information, Bethesda, MD, USA) using the BLAST program 33.

Results

As described in Table 1787 HIV-infected patients were included (735/93.4% males), with a median age of 33 years (Q₁–Q₃: 27–40), of which 50.1% were AIDS patients. A third of them (243/30.9%) were on HAART, with a median time of use of 358 days (Q₁–Q₃: 97.5–847.8); the most frequent ARV scheme used was NRTI + NNRTI (142/58.4%). One hundred and fifty-seven patients (from 421 individuals in whom a viral load determination was available) had undetectable levels of HIV-RNA (37.3%). The median lymphocyte CD4⁺ count, from data available in 439 patients, was 281 cells/mm³ (Q₁–Q₃:140–471. The main transmission category was men who had sex with men (MSM) (562/82.8%), and 52% were current smokers.

The demographic and clinical characteristics of patients, with or without HPV-OL, are presented in Table 1. Fifty-five (6.9%) of the 787 patients showed HPV-OLs (two females and 53 males); 52 of the 53 males (94.5%) were MSM. The patients with HPV-OLs were significantly older (40 years [Q₁–Q₃: 32–44]) than those without lesions (33 years [Q₁–Q₃: 27–39]) (P < 0.001). Also, the subjects with HPV-OLs were in more advanced stages of HIV infection than those in the non-HPV-OLs group (P < 0.023). More patients in the group with HPV-OLs were on HAART in comparison with the other set of patients (74.5% vs. 27.6%, P < 0.001), with NRTIs + NNRTIs being the most frequent HAART combination prescribed in both groups. The duration of HAART use was longer in the HPV-OL group than in patients without HPV-OL (477 vs. 327 days, P = 0.031). The proportion of individuals with an undetectable viral load was significantly higher among the HPV-OLs patients (P < 0.001), but the median CD4⁺ cell counts was comparable in both groups.

The univariate analysis described in Table 2 reveals that the patients on HAART were more likely (OR = 7.7) to present HPV-OLs (95% CI 4.10–14.39, P < 0.001). Also, those with more than 12 months under HAART were 5.1 times as likely to present an HPV-OL (CI 95% 2.84–9.03, P < 0.001) than those with a shorter time on HAART. Other variables significantly associated with the presence of these lesions were age (>40 years old) and undetectable viral load. To explore the independent effect of the variables that were significantly associated in the univariate analysis, a multivariate analysis indicated that the group with HPV-OLs was independently associated with age (≥40 years) and longer time of HAART use (≥12 months).

Of the 55 HPV-OLs, 22 (40%) were squamous cell papilloma (SCP), 15 (27.3%) multifocal epithelial hyperplasia (MEH), and three (5.5%) verruca vulgaris (VV). In seven patients (12.7%), the lesions shared clinical and histopathological characteristics of SCP and VV, so they were defined as oral warts (OW). Eight patients (14.5%) showed concurrent MEH and OW lesions. Oral condyloma acuminatum cases were not found in this study.

The most frequent HPV-OL location was the labial mucosa in 31 (56.4%) cases, followed by the tongue in nine (16.4%), and the buccal mucosa in eight (14.5%) cases. Other affected sites were the palate (n = 3) and the gingiva (n = 3) (5.5% each one). One lesion was found in the floor of the mouth (1.8%). Twenty-six (47.3%) patients presented HPV-OLs in multiple sites on the oral mucosa.

DNA was purified and amplified from 30 HPV-OLs. In 25 patients, it was not possible to obtain a tissue sample because (i) the patient did not give his consent to the procedure, (ii) lesions were small, flat, and did not cause any discomfort, and (iii) patient's systemic conditions were not favorable to perform a surgical procedure (thrombocytopenia). All 30 samples were positive for a β-globin PCR reaction, but there was insufficient DNA to determine the genotype in four biopsy samples. As shown in Table 3, of the 26 included samples, high-risk HPV types (16 and 31) were detected in three (11.5%) and low-risk (types 1, 6, 11, 13, 32, and 74) were found in 17 (65.4%) samples. Two lesions (7.7%) showed mixed high- 16, 18 and low-risk HPV 13 infection, and multiple HPV infection was found in four (15.4%) samples. Human papillomavirus type 13 was the most frequent type identified in 10 patients (38.5%), two of whom also showed HR-HPV (one HPV-16 and one HPV-18). HPV-32 was identified in two MEH samples (7.7%). Cutaneous HPV types (HPV-1, 6, 11) were identified in six samples (one MEH+OW, one VV, two OW, and two SCP).

Discussion

In this study, we found that from 2004 to 2011, 6.9% of 787 HIV/AIDS patients seen at a referral center in Mexico City exhibited HPV-OL. These lesions were associated with older age (>40-years old) and with a longer time on HAART; moreover, patients with more than 12 months under HAART showed a threefold risk of having an HPV-OL. The observed prevalence was higher than the 0.4% reported by our working group for the period 1999–2001 34, and is consistent with the increased trend of HPV-OL in HIV/AIDS patients described by other authors 1-3, 18.

The study by Patton et al. 3 was the first to report that human papillomavirus lesions increased among HIV-positive patients from 2.2% in 1995–1996 to 4.0% in 1996–1999, although the difference did not reach statistical significance. Another study described the increase of oral warts from 5% in individuals without ARVs, to 15% in patients on antiretroviral therapy other than HAART, and to 23% in those on HAART 2. Similarly, Greenwood et al. 1 showed an increase of the prevalence of HPV-OL over time (1996–1998), mainly in patients taking ARVs (4.6%). In addition, King et al. 18 noted an increase in the cumulative incidence of oral warts in a 3-year study period. In recent years, although there is some information regarding the prevalence of HPV-OL in HIV/AIDS patients derived from cross-sectional studies varying from 0.5% 35 to 4.6% 36, there are no longitudinal studies that could bring additional evidence related to the increase of HPV-OL in patients on HAART.

Even though the aforementioned reports 1-3, 18 proposed that HAART might be an important factor that influences the development of HPV-OL, to date, there is not enough scientific evidence to conclude that this therapy could be a risk factor for the development and progression of these lesions.

It is important to highlight that in our findings the development of HPV-OL was associated with older age and a longer duration of HAART use. Thus, the rise of HPV-OL in treated HIV patients may be related to the fact that HAART is an enduring therapy that offers extended survival to HIV patients. It is probable that the long-lasting immunosuppression in HIV-infected individuals placed on HAART has already exceeded some critical threshold for developing HPV-related disease that cannot be reversed with therapy 6. HPV-specific immunity may not recover fully after the immune response is restored, which may explain the relatively limited beneficial effect of HAART on HPV-OL 37. To explain the rise of HPV-OL in the HAART era, another aspect to be considered may be the proliferative capacity of keratinocytes provided by HIV-tat and other HIV genes (rev and vpr) that stimulate the expression of HPV genes (L1, E1, and E6) 16, 38.

In this study, patients with an undetectable viral load had a sixfold risk of presenting HPV-OL, which was expected considering the long time under HAART spent by the HPV-OL individuals. An earlier report 18 described an association between the presence of HPV-OL and a drop of at least one log in the peripheral HIV viral load in the previous 6 months, proposing that oral warts may represent a form of immune reconstitution inflammatory syndrome (IRIS) occurring in response to improved cell-mediated immune function 18. Likewise, a recent review 39 suggested that HPV disease in HIV infection after HAART might represent a persistent immunodysregulation, and a form of immune reconstitution-associated disease (IRAD), a term that also encompasses IRIS.

According to the IRIS Definition criteria proposed by the AIDS Clinical Trial Group 40, the onset of new skin/genital warty lesions, as well as the rapid growth of previously stable warts has been observed during immune restoration 41-45. In addition, in a subset of HIV-infected patients, the clinical appearance of HPV-OL is characterized by a florid onset, with a rapid and uncontrolled course, refractory to conventional treatments 5, 46-49. Nevertheless, a tissue analysis performed in 14 oral warts revealed that HAART had no demonstrable effect on local immune responsiveness; consequently, in agreement with their results, oral warts could not be classified as IRIS 50. Future studies, performed in compliance with international standards for IRIS case definition, may be helpful in clarifying the role of HPV-OL as part of IRIS.

The most frequent HPV-OL identified in this study was SCP (40%), followed by MEH (27.3%), in agreement with recent case reports of SCP 15, 48 and MEH 14, 17, 51 in HIV patients. Also, oral condyloma acuminatum, which has been reported elsewhere 10, 11, 13, 49, was not found in this report.

HPV-OL biopsies from HIV-AIDS individuals have demonstrated a range of low- and high-risk HPV genotypes (HPV-6, 11, 13, 32, 16, 18) 10, 14-16. In this study, the most frequent type was 13, an HPV type present in the oral mucosa that, along with HPV-32, has been associated with MEH in HIV-positive 14, 16 and negative subjects 52-54. We also identified HPV types 1 and 16 in MEH lesions, in agreement with studies in non-HIV-infected people that have isolated HPV 1, 6, 11, 16, and 18 in this type of lesion 52, 53, 55, 56.

Although the majority of studies have emphasized the benign nature of HPV-OL in HIV 17, 28, 46, our observation that a relatively high proportion of sequencing samples showed HR-HPV types (19.2%) underlines the potential for a malignant change in these lesions, particularly considering the increased risk of HPV-associated malignancy among HIV patients on HAART 25.

An interesting finding in this study was the presence of HPV multi-infection in four HPV-OL samples (7.3%). The presence of multiple infections may be associated with the inability of the severely impaired cell-mediated immune response to clear HPV infection in HIV patients 57. On the other hand, even though we cannot differentiate between transient, persistent, or reactivated latent infections, it is necessary to point out the possibility of transient infections, particularly in the set of immunocompromised individuals 16. Also, it has been recognized that sexual behavior is a risk factor for the presence of oral HPV in HIV infection 8, 15, 58-60.

This study has some limitations because it was conducted as a cross-sectional study, so it lacks information on the behavior or progression of HPV-OL over time. Longitudinal studies should be carried out to clarify the possibility of malignant changes in the presence of HR-HPV types. Nevertheless, it is important to mention that the detection of high-risk sequences in tissues does not necessarily prove a causal association with malignancy, thereby reinforcing the high frequency of oral HPV infection in the healthy oral mucosa of HIV patients 61.

Even though, in this study, oral HPV infection was not assessed, it is important to highlight that, in HIV-patients; oral HPV infection frequency could be as high as 80% 61. Interestingly, despite the high prevalence of oral HPV infection found in these studies, HPV-OL have not been described 57, 61-63. Some of the risk factors associated with oral HPV infection, include tobacco consumption, immunological factors, and sexual behavior 8, 63, 64.

In conclusion, it is suggested that the increasing incidence of HPV-OL in the post-HAART era may be associated with the longer life expectancy of individuals who also have an impaired immune system rather than a direct effect of HAART. In HIV-infected patients, the high frequency of HR-HPV infection, added to their increased susceptibility to developing malignancies and the rise in oropharyngeal HPV-associated cancer emphasizes the need for adequate screening programs and timely interventions in HIV-infected patients.

Conflict of interest

None declared.