Noninvasive Sampling Reveals Distinct Human Papillomavirus and Immune Signatures Associated With Cytological Abnormalities in People Living With HIV-1

Abstract

Background

Persistent high-risk (HR) human papillomavirus (HPV) infections drive most anal cancers, disproportionately affecting people with human immunodeficiency virus (PLH), for whom cytology-based screening remains essential. However, the biological context of abnormal anal cytological findings, particular regarding local immune cells, is incompletely understood.

Methods

In this pilot study, 57 PLH underwent longitudinal noninvasive anal sampling over 1 year, including cytology, HPV genotyping, and phenotypic characterization of anomucosal immune cells.

Results

Anal HPV infection and persistence were highly prevalent. Cytological abnormalities were associated with more concurrent (HR) HPV infections, lower viral loads, and anal immune microenvironment characterized by increased CD4 and CD8 T cells with up-regulated tissue residency, activation, and effector markers.

Conclusions

Noninvasive anal sampling revealed that abnormal anal cytological findings in PLH are characterized by a distinct microenvironmental signature marked by multiple (HR) HPV infections and increased levels of tissue resident, activated, and effector anomucosal T cells, potentially highlighting the important role of local immune cells.

Human papillomavirus (HPV) is the cause of multiple anogenital and oropharyngeal epithelial cancers [1]. HPV types differ markedly in their oncogenic potential. While low-risk (LR) types primarily cause anogenital warts and are rarely associated with cancer, persistent infections with high-risk (HR) types account for about 90% of anal cancers. Putative HR types are only rarely linked to cancer, typically under specific conditions such as immunosuppression [2]. Among HR types, HPV-16 is considered the most cancerogenic, especially for anal cancer, likely due to its pronounced immune-evasive properties [1]. In addition to risk factors like tobacco smoking, exposure to HPV through receptive anal intercourse and immune impairments—particularly human immunodeficiency virus 1 (hereafter HIV) infection—substantially increase anal cancer risk. Incidence rates are markedly elevated in men living with HIV regardless of sexual orientation and is highest among men who have sex with men (MSM) and are living with HIV [3].

People living with HIV (PLH) are at higher risk at every stage of HPV-associated cancer development, from HPV infection to faster progression to precancerous lesions [3]. Thus guideline-recommended cytology screening in this patient cohort for the early detection of (pre)malignant lesions is an essential yet (due to the lack of specificity) suboptimal part of routine checkups, especially in the growing population of aging PLH [4].

Spontaneous regression of HPV-associated premalignant lesions has been associated with systemic T-cell responses [5]. Yet much less is known about immune processes at the site of infection—the anal mucosa. Although T cells at the site of infection are considered crucial for local immune control [6], studies analyzing these cells in the context of HPV-associated lesions and abnormal cytological findings remain scarce [7]. In this pilot study, we aimed to gain insight into the immune microenvironment associated with cytological abnormalities in PLH by applying an innovative, noninvasive anal sampling approach to access anomucosal T cells.

METHODS

Detailed information on reagents and devices is provided in Supplementary Tables 1 and 2.

Study Cohort and Sampling

PLH attending the outpatient clinic at Ludwig-Maximilians-University Munich (LMU) Hospital were invited to participate in our observational and prospective follow-up study (approved by the Institutional Review Board of Medical Faculty at LMU [19-0849]). Besides the requirement of ≥1 evaluable anal cytological result during the study period, no additional inclusion criteria were set. After written informed consent was obtained from 57 study participants, samples were collected at baseline and during follow-up visits for up to 1 year (week 13 [interquartile range, weeks 13–15], week 27 [26–28], and week 53 [51–55]).

At each visit, anal swab samples were collected as follows: First, as part of routine care, PLH were offered annual anal cancer screening with a cytology smear, which was evaluated by a trained cytopathologist according to the Bethesda System. Second, an anal swab sample was collected to isolate anal (immune) cells, followed by an additional swab sampled for HPV genotyping. All swabs were collected with study participants lying down in the left lateral position with hips flexed, after which a swab was inserted 2–3 cm into the anal canal, rotated with gentle pressure, and slowly withdrawn. The brush for cytology screening was rolled out on a microscope slide, and cells were fixed. Despite being recommended by clinical guidelines, histopathological follow-up of abnormal cytological findings was inadequately performed in this center and for this study cohort (as analyzed and discussed elsewhere [4]), which prevented its inclusion in the present analysis.

The swab samples for DNA extraction were put into 3-mL ThinPrep Preserve-Cyt solution or in 5-mL R10 medium for cell isolation and vortexed immediately. To isolate immune cells but remove epithelial cells, the swab was removed from R10 medium and washed with phosphate-buffered saline over a 100-µm cell strainer. Cells in the medium were pelleted, resuspended in phosphate-buffered saline, filtered through the cell strainer, and cryopreserved (90% fetal calf serum + 10% dimethyl sulfoxide) in liquid nitrogen. The cytopathologist was blinded to all other study data, but the laboratory personnel were not.

Anal HPV Genotyping

DNA isolated from 500-µL ThinPrep Preserve-Cyt solution with QIAamp DNA Mini Kit was analyzed with the Allplex HPV28 Detection Kit identifying LR (HPV types 6, 11, 40, 42, 43, 44, 54, and 61), putative HR (including probably and possibly cancerogenic types: types 26, 53, 66, 68, 69, 70, 73, and 82), and HR HPV types (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59), as classified by the World Health Organization [2, 8]. Both kits were used following the manufacturers’ protocols. Viral loads were calculated as 1/normalized cycle threshold value, which were normalized to a housekeeping gene. Clearance was defined as absence of a baseline-detected type in 2 consecutive follow-ups within 1 year. Persistence was defined as detection of the same specific type at baseline and at 2 consecutive follow-up visits within 1 year.

Phenotyping of Anomucosal Immune Cells

To characterize anal immune status, cells from swab samples were phenotyped by means of flow cytometry. Thawed cells were washed, stained for viability with the ZombieNIR Fixable Viability Kit (15 minutes at room temperature [RT]), and incubated with fluorochrome-conjugated antibodies in a master mix with Brilliant Stain Buffer (20 minutes at RT). After being washed with fluorescence-activated cell sorting buffer, cells were fixed with Fix&Perm Medium A (15 minutes at RT) and washed again. Cells were filtered with a 35-µm cell strainer before acquisition on a LSRFortessa Cell Analyzer (BD Bioscience) on the same day. Rainbow beads assured interexperimental consistency. Data were analyzed using FlowJo software (version 10.10). Supplementary Figure 1 shows the exemplary gating of single-positive cells for phenotypic markers on T cells (lymphocytes > singlets > live cells > CD3⁺CD45⁺ > CD4⁺ or CD8⁺). Double-positive cells (PD1⁺HLA-DR⁺ or CD69⁺CD103⁺) were identified during analysis by using Boolean gates. Only samples with >20 CD4 or CD8 T cells were included, allowing gating of downstream markers with simultaneous inclusion of a sufficient number of samples and resulting in 32 evaluable samples for CD4 or CD8 T cells.

Statistics

P values were calculated with GraphPad Prism software (version 10.1.2), and χ² or Mann-Whitney tests were used for 2-group comparisons where applicable. Effect sizes and 95% confidence intervals were calculated with JASP 0.95.4.0 software. Differences were considered statistically significant at P < .05. Data are expressed as medians with interquartile ranges, except where otherwise specified.

RESULTS

The majority of participants in our cohort were male (54 of 57 [95%]), of whom 44 identified as MSM. Their median age was 54 years. All participants PWH taking suppressive antiretroviral therapy (see Supplementary Table 3 for cohort characteristics). At the baseline visit, the prevalence of anal HPV infection was high, with 51 of 57 participants (89%) carrying ≥1 HPV type (Supplementary Table 4). Most individuals (40 of 57 [70%]) carried an HR type, with HPV-16 the most prevalent (11 of 57 [19%]). In addition, accompanying LR or probably and possibly cancerogenic types were detected (a median of 3 types). After 1 year, HR HPV infections frequently persisted (in 45 of 54 participants [83%] [Supplementary Table 4], with detection of the same specific type at baseline and 2 consecutive follow-up visits). On the other hand, LR infections were more frequently cleared within the 1-year follow-up period (with clearance defined as the absence of a baseline-detected type in 2 consecutive follow-ups) (Supplementary Table 4).

Participants without and those with cytological abnormalities (19 of 57 [33%]) (Figure 1A) did not differ significantly with respect to established risk factors for anal cancer, such as low CD4 nadir count or smoking status (Supplementary Table 3). However, an abnormal cytological result was associated with lower viral loads (defined as 1/normalized cycle threshold value) of HR HPV types, which are considered the causative agent of (pre)malignant lesions, and a higher number of concurrent infections, including HR types overall and specifically HPV-56 and HPV-51 (Figure 1B–1D).

Figure 1.

Anal human papillomavirus (HPV) burden and type distribution according to cytological results. A, Anal cytological results collected from routine clinical samples. Abnormal cytological findings include atypical squamous cells of undetermined significance (ASC-US [light blue]; n = 8), low-grade squamous intraepithelial lesion (LSIL [medium blue]; n = 7), and high-grade squamous intraepithelial lesion (HSIL [dark blue]; n = 4). Abbreviation: NILM, negative for intraepithelial lesion and malignancy. B–D, Comparison between patients negative for intraepithelial lesions and malignancy (NLIM [gray]) and individuals with cytological abnormalities (blue). Anal HPV status was assessed with Allplex HPV28 Detection Kit detecting low-risk (LR; HPV types 6, 11, 40, 42, 43, 44, 54, 61), probably and possibly high-risk (HR) (types 26, 53, 66, 68, 69, 70, 73, 82) or HR (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59) HPV types. (One participant refused to participate in >2 study visits, so data for this individual is missing.) B, Viral load of all HR-HPV types detected throughout the study period at 3 study visits. Viral load is defined as 1/normalized cycle threshold (Ct) values, which were normalized to a housekeeping gene. HPV-16 infections are highlighted in red. P values (displayed above graphs throughout figure) were calculated using the Mann-Whitney test. C, Number of all detected types (left) and detected HR types (right) per patient throughout the study period at 3 study visits. P values were calculated using the Mann-Whitney test; for effect sizes and confidence intervals, see Supplementary Table 5. D, HR type distribution. P values were calculated using χ² tests, based on absolute numbers.

Using our noninvasive sampling technique, we phenotyped T cells from anal swab samples collected throughout the study. While HR-HPV infection did not alter anomucosal T-cell counts or phenotypes (Supplementary Figure 2), abnormal cytological results were reflected by features indicating T-cell recruitment. The anal microenvironment was characterized by an increased number of CD4 (not significant) and CD8 (P = .02; r_rb = 0.469 [95% confidence interval, .102–.723]) T cells (Figure 2A and 2B), which exhibited phenotypic markers associated with tissue residency, activation, and effector functions (Figure 2C and 2D). Markers of tissue-resident cells defined by CD103⁺CD69⁺ coexpression as well as mucosal homing (α4β7) were particularly prominent on CD8 T cells in the context of cytological abnormalities. Moreover, CD4 T-cell–coexpressing HLA-DR/PD1 were significantly more pronounced in samples with abnormal cytological findings.

Figure 2.

Abnormal anal cytological findings are associated with a T-cell–enriched microenvironment. A, Anal cells collected with a swab were analyzed by means of flow cytometry. Detailed gating is depicted in Supplementary Figure 1. Abbreviations: FSC, forward scatter; SSC, side scatter. B, Cell counts of CD4 and CD8 T cells. C, D, Their coexpression with tissue homing (CD69, CD103, and α4β7), activation (PD1 and HLA-DR), and effector function (chemokine receptor CCR5) markers in patients negative for intraepithelial lesion and malignancy (NILM [gray]) or with abnormal cytological results (blue). A–D, All available samples were measured, but only samples with >20 CD4 or CD8 T cells were included in the respective analyses. P values (displayed above graphs throughout figure) were calculated using the Mann-Whitney-test; for effect sizes and confidence intervals, refer to Supplementary Table 5. Abbreviation: NS, not significant.

DISCUSSION

This study aimed to improve our understanding of microenvironmental changes associated with cytological abnormalities by investigating the HPV burden and the composition of the anomucosal immune cell compartment. Distinguishing between transient and persistent HR HPV infection—with the latter being more relevant for lesion development—is particularly important in individuals with multiple sexual partners, in whom transient infections and reinfections are common [9].

In our cohort of PLH, however, both the baseline burden of HPV infection and the frequency of persistent infection over the 1-year follow-up period were strikingly high. This pattern suggests a great overall exposure to HPV and, over time, a potentially increased risk of developing (pre)cancerous lesions. Notably, individuals with abnormal cytological findings in our pilot cohort harbored a significantly higher number of concurrent (HR) HPV types and—maybe due to that—a lower per-virus viral load. Although HPV-51 and HPV-56 are significantly linked to abnormal cytological findings in our cohort, their association with cancer progression has been reported low [10]. In contrast, HPV-16—the most oncogenic type—did not differ significantly between the 2 groups. This may reflect the fact that most cytological abnormalities detected in our study were at an early stage, as previous studies have shown that the prevalence of HPV increases with disease severity [10].

Using our innovative sampling technique, we next identified a T-cell–enriched microenvironment associated with cytological abnormalities in our observational study. However, it needs to be stated that only approximately half of anal T-cell samples could be included due to fecal contamination or low T-cell yield. Despite capturing comparable small T-cell numbers, which might avoid more complex downstream application (eg, functional assays), this minimally invasive, swab-based sampling enables easier and less harmful sample collection compared with biopsies and thus might help improve the limited understanding of local immune cells.

However, considering the modest sample size and limited number of evaluable immune-cell samples, our immunological findings should be viewed as exploratory observations that generate hypotheses for future studies rather than providing confirmatory evidence. This technology revealed a distinct immune profile regarding abundance and phenotypes of anomucosal T cells in individuals with abnormal cytological findings. In particular, CD4 T cells coexpressing HLA-DR/PD1 were more abundant in these samples. While PD1 expression can indicate exhaustion, the concurrent expression of PD1 and HLA-DR—along with increased effector CD4 T cells—suggests immune activation, analogously to findings in cervical samples with cytological abnormalities [11]. Accordingly, HPV-associated abnormalities in the anogenital tract do not appear to be “immunologically quiescent.” In addition, we observed high numbers of the functional heterogenous CD8 tissue-resident cell population, which is essential in controlling persistent viral infections and has been associated with an improved outcome in multiple solid cancers [12].

The potential underlying cause and functional role of the T-cell–enriched microenvironment remains unknown. As speculation, local immune cells might be involved in lesion regression, as suggested by the successful treatment of HPV-associated warts with the immunactivator Imiquimod [13] and improved outcomes in HPV-related abnormalities in the context of local inflammation [14]. However, sexual behavior and hygiene practices in our predominantly MSM cohort might also contribute to local inflammation [15]. As recently shown, a proinflammatory milieu might negatively influence the specificity of anal cytology, which is generally reported to be relatively low [4]. Accordingly, limitations of the current study include the lack of screening for concomitant sexually transmitted infections, which may further promote a proinflammatory milieu, thereby potentially reducing the specificity of anal cytology. Another limitation is that group comparisons were based on cytological findings, as no histopathological follow-up was available. This also reflects the limited protocol adherence observed in routine clinical practice [4]. Further research defining the anal immune microenvironment will be required to better understand the role of the immune infiltrate. Such studies should also include an appropriate comparator group, for example, MSM receiving HIV preexposure prophylaxis.

Furthermore, for a pilot study, we have chosen a relatively limited sample size and a 1-year follow-up period, which might lead to an overinterpretation of persistence, as some types may have cleared with longer observation. In addition, while we have data for our cohort on factors including smoking status, CD4 count, and antiretroviral therapy duration, the small sample size did not allow for multivariate modeling.

In summary, using an innovative noninvasive sampling method, we report that abnormal anal cytological findings in our observational pilot study of 57 PLH is characterized by a distinct microenvironmental signature marked by multiple (HR) HPV infections and increased levels of tissue resident, activated, and effector anomucosal T cells, potentially highlighting the important role of local immune cells.

Supplementary Data

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.