Abstract
As the US moves increasingly towards using human papillomavirus (HPV) testing with or without concurrent cytology for cervical cancer screening, it is unknown what the corresponding risks are following a screening result for women living with HIV (WLWH), which will dictate the optimal clinical follow-up. Therefore, using medical records data from Kaiser Permanente Northern California, which introduced triennial HPV and cytology co-testing in women aged 30–64 years in 2003, we compared risks of cervical intraepithelial neoplasia grade 2 (CIN2) or more severe diagnoses (CIN2+) in women not known to have HIV (HIV[−] women) (n=67,488) frequency matched 111:1 on age and year of the first co-test to the 608 WLWH (n=608). WLWH were more likely to test HPV positive (20.2% vs. 6.5%, p<0.001) and have non-normal cytology (14.1% vs. 4.1%, p<0.001) than HIV[−] women. Five-year CIN2+ risks for all WLWH and HIV[−] women were 3.5% (95%CI=2.0–5.0%) and 1.6% (95%CI=1.5–1.8%) (p=0.01), respectively. Five-year CIN2+ risks for WLWH with positive HPV and non-normal cytology, positive HPV and normal cytology, negative HPV and non-normal cytology, and negative HPV and normal cytology were 24.9% (95%CI=13.4–36.4%), 3.0% (95%CI=0.0–7.4%), 3.6 (95%CI=0.0–9.8%) and 0.3% (95%CI=0.0–0.8%), respectively. Corresponding 5-year CIN2+ risks for HIV[−] women were 26.6% (95%CI=24.6–28.7%), 8.5% (95%CI=7.2–9.9%), 1.9% (95%CI=1.0–2.8%), and 0.5% (95%CI=0.4–0.6%), respectively. Thus, in this healthcare setting, the main cause in overall CIN2+ risk differences between WLWH and HIV[−] women was the former was more likely to screen positive and once the screening result is known, it may be reasonable to manage both populations similarly.
Keywords: Human papillomavirus (HPV), cytology, Pap, cervical intraepithelial neoplasia, cervical cancer, human immunodeficiency virus (HIV)
Introduction
Current cervical-cancer screening recommendations are (Curry et al., 2018): 1) triennial cytology screening for women aged 21–64 years, and 2) human papillomavirus (HPV) testing alone or concurrently with cytology (“co-testing”) every 5 years or triennial cervical cytology screening for women aged 30–64 years. Due to their greater overall cervical-cancer risk (Grulich et al., 2007; Silverberg and al, 2015; Stelzle et al., 2021), separate guidelines have been establish for women living with human immunodeficiency virus (HIV) (WLWH) (2021): 1) annual cytology, and following 3 negative results, triennial cytology for WLWH aged 21–29 years, and 2) triennial co-testing or annual cytology, and following 3 negative results, triennial cytology for WLWH aged ≥30 years.
There are few data on co-testing of WLWH. One observational cohort reported risks of cervical intraepithelial neoplasia (CIN) grade 2 (CIN2) or more severe diagnoses (CIN2+), stratified by HPV and cytology results, to be similar between WLWH and HIV[−] women (Keller et al., 2015; Keller et al., 2012). However, these results were observational in WLWH who were undergoing routine cervical cytology every 6- to 12-months, do not reflect those outcomes following routine co-testing and care, and the HIV[−] women in that study were selected to be at high risk for HIV. We therefore updated our previous analysis (Castle et al., 2012) to compare cervical outcomes in WLWH to an unselected population of HIV[−] women undergoing routine co-testing in the same managed healthcare system.
Methods
Study Population.
The Kaiser Permanente Northern California (KPNC) cohort consist of approximately 2.3 million women, approximately 1.6 million aged 30–64 years, enrolled in screening from January 1, 2003 to February 21, 2021. This cohort has been extensively described (Katki et al., 2011). KPNC screened women aged 30–64 years by HPV and cytology co-testing (and women aged 21–29 with cytology alone) until 2013, when it lowered the co-testing screening age to 25 years. The KPNC institutional review board approved use of the data, and National Institutes of Health Office of Human Subjects Research deemed this study exempt from review.
Screening and Clinical Management.
Women were screened by HPV and Pap/cervical cytology co-testing as previously described (Gage et al., 2014; Katki et al., 2011). Women with abnormal screening tests and diagnoses were managed generally in accordance with national recommendations during that period.
Statistical methods.
To estimate relative risks, we frequency matched women not known to have HIV (HIV[−] women) to WLWH on exact year of age and calendar year of first co-test. Both relative and absolute risks of CIN2+ or CIN grade 3 or more severe diagnoses (CIN3+) were estimated using prevalence-incidence mixture models (Cheung et al., 2017; Hyun et al., 2017). These models combine a logistic regression model for prevalent high-grade cervical abnormalities and a proportional hazards model for incident high-grade cervical abnormalities, while accounting for delayed detection of prevalent high-grade cervical abnormalities due to colposcopy protocols (left censoring) and time of onset of incident high-grade cervical abnormalities occurring between two assessment visits (interval censoring). Women who are not cases (detected with CIN2+ or CIN3+ according to the analysis) were right censored at the time of their last negative assessment visit, which is defined as having <CIN2 histology or having a negative co-test result. Women who were not cases and had no negative assessment visit did not contribute to risk estimation (i.e., they are non-informative for the maximum likelihood estimation). An example of an individual not used to estimate risk is a woman testing HPV+/LSIL that did not attend colposcopy or have further follow-up visits. For women who had their first co-test less than 5 years from the final date of this dataset, they contributed to the risk estimates as either cases or right-censored controls. Even without follow-up beyond one visit (plus colposcopy, if applicable), they contribute to the immediate risk estimate, which is part of the 5-year cumulative risk estimate.
Both relative and absolute risk estimates of high-grade cervical abnormalities associated with HIV were adjusted for HPV status and cytology. We considered HPV-positive atypical squamous cells of undetermined significance, low-grade squamous intraepithelial lesion, or more severe cytologic interpretations as a positive (non-normal) cytology.
We calculated cumulative incidence (risk) of high-grade cervical abnormalities out to 5 years for the matched cohorts adjusting for HPV and cytology, then calculated weighted risk overall and stratified on HPV and cytology results. Although presented absolute risks for HIV-negative were estimated from the subset of matched controls, results agree with risks estimated from the full HIV-negative KPNC population (data not shown). Due to lack of CIN3+ cases among HPV-negative WLWH, we subsetted those models by HPV status to estimate the HPV-positive cumulative risk. The associated lack of standard errors prohibited us from calculating weighted risk estimates, instead we report marginal results in the overall and stratified results. All analysis were run in R 4.1.2. P of <0.05 was considered statistically significant.
Results
Seven hundred sixty-eight WLWH were identified, of which 674 (87.8%) had follow-up. Among those with follow-up, 556 (82.5%) had prevalent HIV at the time of the first co-test identified. Another 118 (17.5%) women were diagnosed with HIV after they became KPNC members, and their index co-test was the co-test after becoming HIV positive. Six hundred eight (90.8%) were in the age range of 30–64 years for routine screening by co-testing and were included in this analysis. HIV[−] women were matched 111:1 on age and year of the first co-test to the 608 WLWH, generating an HIV[−] sample of 67,488 women.
The median and mean age at time at index screen was 44 and 44.56 years, respectively. The distribution of year of first co-test was 20.2% for 2003–7, 28.6% for 2008–12, 31.9% for 2013–17, and 19.2% for 2018 or later. There was a significant difference in the racial composition of the two (p<0.001), with black being the most common race among WLWH (40.8%) and white being the most common race among the HIV[−] women (34.8%) (Table 1). WLWH were more likely than HIV[−] women to test HPV positive (20.2% vs. 6.5%, respectively, p<0.001) and have non-normal cytology (14.1% vs. 4.1%, respectively, p<0.001). WLWH also had many more visits than HIV[−] (p<0.001), which could not be explained by screening results (data not shown).
Table 1.
Characteristics of study populations of women living with human immunodeficiency virus (HIV) (WLWH) matched with those who were not (HIV[−] women).
| HIV[−] Women (n=67,488) | WLWH (n=608) | ||||
|---|---|---|---|---|---|
| n | % | n | % | P | |
| Race | |||||
| Missing/Unknown | 18,727 | 27.7 | 58 | 9.5 | <0.001 |
| White | 23,489 | 34.8 | 160 | 26.3 | |
| Black | 3,976 | 6.9 | 248 | 40.8 | |
| Asian/Pacific Islander | 10,911 | 16.2 | 51 | 8.4 | |
| Hispanic | 9,224 | 13.7 | 73 | 12.1 | |
| Native American | 273 | 0.4 | 4 | 0.7 | |
| Other | 888 | 1.3 | 14 | 2.3 | |
| HPV Results | |||||
| Positive | 4,355 | 6.5 | 123 | 20.2 | <0.001 |
| Negative | 6,1966 | 91.8 | 475 | 78.1 | |
| Missing | 11,167 | 1.7 | 10 | 1.6 | |
| Cytology Results | |||||
| Non-normal | 2,615 | 3.9 | 84 | 13.8 | <0.001 |
| Normal* | 64,713 | 95.9 | 522 | 85.9 | |
| Missing | 160 | 0.2 | 2 | 0.3 | |
| High-Grade** | 547 | 0.8 | 9 | 1.5 | <0.001 |
| Low-Grade | 2,068 | 3.1 | 75 | 12.3 | |
| Negative* | 64,713 | 95.9 | 522 | 85.9 | |
| Missing | 160 | 0.2 | 2 | 0.3 | |
| Number of follow-up visits | |||||
| 1 | 28,834 | 42.7 | 146 | 24.0 | <0.001 |
| 2 | 15,173 | 22.5 | 100 | 16.4 | |
| 3 | 8,972 | 13.2 | 78 | 12.8 | |
| 4 | 6,775 | 10.0 | 63 | 10.4 | |
| 5 | 4,301 | 6.4 | 70 | 11.5 | |
| ≥6 | 3,433 | 5.1 | 151 | 24.8 | |
Included atypical squamous cells of undetermined significance if the concurrent HPV test was negative.
Included cancer, high-grade squamous intraepithelial lesion (HSIL, atypical squamous cells cannot rule out HSIL, and atypical glandular cells.
CIN2+ and CIN3+ risks by HIV status and baseline screening results are shown in Table 2. Five-year CIN2+ risks for all WLWH and HIV[−] women were 3.5% (95%CI=2.0–5.0%) and 1.6% (95%CI=1.5–1.8%) (p=0.01), respectively. Five-year cumulative CIN3+ risks for all WLWH and HIV[−] women were 1.3% (95%CI=0.3–2.3%) and 0.7 (95%CI=0.6–038%) (p=0.24), respectively.
Table 2.
A comparison of cervical intraepithelial neoplasia grade 2 or more severe diagnoses (CIN2+) or grade 3 or more severe diagnoses (CIN3+) risks with 95% confidence intervals (95%CI) in women living with HIV (WLWH) and without HIV (HIV[−]), stratified on human papillomavirus (HPV) and cytology co-testing results. The first two row are the total population of WLWH and HIV[−] women. Risk estimates for paired WLWH and HIV[−] groups in which there were fewer than 30 WLWH and no cases of CIN2+ for those WLWH are not included. “Any” is for any result e.g., positive (Pos), negative (Neg), or not available (n/a).
| HIV Status | HPV Result | Cytology Result | NTotal | Ninformative** | End point | NCases | Immediate (Prevalent) | 1-Year Cumulative Risk | 3-Year Cumulative Risk | 5-Year Cumulative Risk | pCIN2+†╱pCIN3+‡ | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Risk (%) | 95%CI (%) | Risk (%) | 95%CI (%) | Risk (%) | 95%CI (%) | Risk (%) | 95%CI (%) | ╱ | |||||||
| WLWH | Any | Any | 608 | 474 | CIN2+ | 22 | 2.4 | 1.2–3.5 | 2.6 | 1.4–3.9 | 3.1 | 1.8–4.4 | 3.5 | 2.0–5.0 | 0.01╱0.24 |
| CIN3+§ | 7 | 0.8 | 0.0–1.6 | 0.9 | 0.0–1.7 | 1.0 | 0.2–1.8 | 1.3 | 0.3–2.3 | ||||||
| HIV [−] | 67,488 | 39,398 | CIN2+ | 907 | 0.8 | 0.7–0.9 | 1.2 | 1.1–1.3 | 1.4 | 1.3–1.5 | 1.6 | 1.5–1.8 | |||
| CIN3+ | 377 | 0.4 | 0.4–0.5 | 0.5 | 0.5–0.6 | 0.7 | 0.6–0.7 | 0.7 | 0.6–0.8 | ||||||
| WLWH | Pos | Any | 123 | 111 | CIN2+ | 20 | 10.5 | 5.1–16 | 11.9 | 6.2–17.6 | 13.4 | 7.4–19.5 | 15.6 | 8.7–22.5 | 0.53╱0.38 |
| CIN3+ | 7 | 3.6 | 0.2–7 | 3.6 | 0.2–7 | 4.3 | 0.6–7.9 | 5.8 | 1.0–10.6 | ||||||
| HIV [−] | 4355 | 3712 | CIN2+ | 665 | 11.2 | 10–12.4 | 13.8 | 12.7–14.8 | 16.2 | 15–17.3 | 17.9 | 16.6–19.1 | |||
| CIN3+ | 298 | 5.4 | 4.5–6.2 | 6.6 | 5.8–7.3 | 7.4 | 6.6–8.2 | 8.0 | 7.1–8.9 | ||||||
| WLWH | Neg | 475 | 355 | CIN2+ | 2 | 0.2 | 0–0.7 | 0.2 | 0–0.7 | 0.4 | 0–0.8 | 0.4 | 0–0.9 | 0.76╱n/a | |
| CIN3+§ | 0 | 0.0 | 0.0–0.8‡ | 0.0 | n/a | 0.0 | n/a | 0.0 | n/a | ||||||
| HIV [−] | 61,966 | 35,640 | CIN2+ | 236 | 0.1 | 0–0.2 | 0.3 | 0.2–0.4 | 0.4 | 0.3–0.5 | 0.5 | 0.4–0.6 | |||
| CIN3+ | 77 | 0.1 | 0.1–0.1 | 0.1 | 0.1–0.1 | 0.2 | 0.2–0.2 | 0.2 | 0.2–0.2 | ||||||
| WLWH | n/a | 10 | 8 | CIN2+ | 0 | ╱ | |||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| HIV [−] | 1152 | 44 | CIN2+ | 6 | ╱ | ||||||||||
| CIN3+ | 2 | ╱ | |||||||||||||
| WLWH | Any | Pos | 84 | 79 | CIN2+ | 19 | 14.4 | 6.7–22.1 | 16.5 | 8.4–24.6 | 18.7 | 10.1–27.4 | 22.0 | 12.0–32.0 | 0.77╱0.11 |
| CIN3+§ | 5 | 3.8 | 0.0–8.0 | 3.8 | 0.0–8.1 | 4.6 | 0.3–8.9 | 6.4 | 1.0–11.8 | ||||||
| HIV [−] | 2615 | 2357 | CIN2+ | 522 | 17.2 | 15.7–18.7 | 19.6 | 18.0–21.2 | 21.9 | 20.2–23.6 | 23.5 | 21.7–25.3 | |||
| CIN3+ | 234 | 8.5 | 7.4–9.7 | 9.6 | 8.4–10.8 | 10.4 | 9.1–11.7 | 11.0 | 9.6–12.3 | ||||||
| WLWH | Neg* | 522 | 394 | CIN2+ | 3 | 0.4 | 0–1.0 | 0.4 | 0–1.0 | 0.6 | 0–1.1 | 0.6 | 0–1.1 | 0.50╱0.43 | |
| CIN3+§ | 2 | 0.3 | 0.0–0.8 | 0.3 | 0.0–0.8 | 0.3 | 0.0–0.8 | 0.4 | 0.0–1.0 | ||||||
| HIV [−] | 64,713 | 36,981 | CIN2+ | 383 | 0.2 | 0.1–0.2 | 0.4 | 0.4–0.5 | 0.6 | 0.5–0.7 | 0.8 | 0.7–0.8 | |||
| CIN3+ | 133 | 0.1 | 0.1–0.2 | 0.2 | 0.1–0.2 | 0.3 | 0.2–0.3 | 0.3 | 0.3–0.3 | ||||||
| WLWH | n/a | 2 | 1 | CIN2+ | 0 | ╱ | |||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| HIV[−] | 145 | 58 | CIN2+ | 2 | ╱ | ||||||||||
| CIN3+ | 1 | ╱ | |||||||||||||
| WLWH | Pos | Pos | 71 | 66 | CIN2+ | 18 | 16.4 | 7.5–25.3 | 18.7 | 9.4–28.1 | 21.2 | 11.3–31.2 | 24.9 | 13.4–36.4 | 0.77╱0.21 |
| CIN3+⁑ | 5 | 4.6 | 0.0–9.6 | 4.6 | 0.0–9.6 | 5.5 | 0.1–10.9 | 7.6 | 0.3–14.9 | ||||||
| HIV [−] | 2,249 | 2,081 | CIN2+ | 494 | 19.5 | 17.8–21.2 | 22.2 | 20.4–24.0 | 24.8 | 22.9–26.7 | 26.6 | 24.6–28.7 | |||
| CIN3+ | 230 | 9.6 | 8.4–10.9 | 10.9 | 9.5–12.2 | 11.8 | 10.4–13.2 | 12.4 | 10.9–13.9 | ||||||
| WLWH | Neg* | 52 | 45 | CIN2+ | 2 | 2.5 | 0.0–6.9 | 2.6 | 0.0–6.5 | 2.8 | 0.0–7.2 | 3.0 | 0.0–7.4 | 0.02╱0.97 | |
| CIN3+⁑ | 2 | 2.2 | 0.0–6.5 | 2.2 | 0.0–6.5 | 2.6 | 0.0–7.0 | 3.4 | 0.0–8.7 | ||||||
| HIV [−] | 2,097 | 1,625 | CIN2+ | 169 | 2.3 | 0.6–3.9 | 4.7 | 3.6–5.7 | 6.9 | 5.8–8.1 | 8.5 | 7.2–9.9 | |||
| CIN3+ | 67 | 0.8 | 0.0–1.9 | 1.9 | 1.2–2.6 | 2.7 | 2.0–3.5 | 3.3 | 2.4–4.2 | ||||||
| WLWH | n/a | 0 | 0 | CIN2+ | 0 | ╱ | |||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| HIV [−] | 9 | 6 | CIN2+ | 2 | ╱ | ||||||||||
| CIN3 + | 1 | ╱ | |||||||||||||
| WLWH | Neg | Pos | 11 | 11 | CIN2+ | 1 | 1.6 | 0.0–5.7 | 2.1 | 0.0–6.4 | 2.7 | 0.0–7.5 | 3.6 | 0.0–9.8 | 0.59╱n/a |
| CIN3+ | 0 | 0.0 | 0.0–33.5‡ | 0.0 | n/a | 0.0 | n/a | 0.0 | n/a | ||||||
| HIV [−] | 321 | 258 | CIN2+ | 23 | 1.4 | 0.4–2.4 | 1.6 | 0.6–2.6 | 1.8 | 0.8–2.7 | 1.9 | 1.0–2.8 | |||
| CIN3+ | 11 | 1.0 | 0.0–2.2 | 1.0 | 0.0–2.2 | 1.1 | 0.0–2.3 | 1.1 | 0.0–2.3 | ||||||
| WLWH | Neg* | 462 | 343 | CIN2+ | 1 | 0.2 | 0.0–0.7 | 0.2 | 0.0–0.7 | 0.3 | 0.0–0.7 | 0.3 | 0.0–0.8 | 0.44╱n/a | |
| CIN3+ | 0 | 0.0 | 0.0–0.8‡ | 0.0 | n/a | 0.0 | n/a | 0.0 | n/a | ||||||
| HIV [−] | 61,509 | 35,330 | CIN2+ | 213 | 0.1 | 0.1–0.2 | 0.3 | 0.2–0.3 | 0.4 | 0.4–0.5 | 0.5 | 0.4–0.6 | |||
| CIN3+ | 66 | 0.1 | 0.0–0.1 | 0.1 | 0.1–0.2 | 0.2 | 0.1–0.2 | 0.2 | 0.1–0.2 | ||||||
| WLWH | n/a | 2 | 1 | CIN2+ | 0 | ╱ | |||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| HIV [−] | 136 | 52 | CIN2+ | 0 | ╱ | ||||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| WLWH | n/a | Pos | 2 | 2 | CIN2+ | 0 | ╱ | ||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| HIV [−] | 45 | 2 | CIN2+ | 0 | ╱ | ||||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| WLWH | Neg | 8 | 6 | CIN2+ | 0 | ╱ | |||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| HIV [−] | 1107 | 26 | CIN2+ | 1 | ╱ | ||||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| WLWH | n/a | 0 | 0 | CIN2+ | 0 | ╱ | |||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
| HIV [−] | 15 | 2 | CIN2+ | 0 | ╱ | ||||||||||
| CIN3+ | 0 | ╱ | |||||||||||||
Included atypical squamous cells of undetermined significance if the concurrent HPV test was negative.
Differences in 5-year risk between WLWH and HIV[−] for CIN2+
Differences in 5-year risk between WLWH and HIV[−] for CIN3+
Poisson exact confidence interval
Modeled in HPV+ subset
Marginal analysis instead of weighted average
Ninformative refers to women who contributed to the risk estimations because they were diagnosed with CIN2+ or had a negative assessment in follow-up.
Five-year CIN2+ risks following a positive HPV test were 15.6% (95%CI=8.7–22.5%) for WLWH and 17.9% (95%=16.6–19.1%) for HIV[−] women (p=0.53). Five-year CIN2+ risks following a negative HPV test were 0.4% (95%CI=0.0–0.9%) for WLWH and 0.5% (95%=0.4–0.6%) for HIV[−] women (p=0.76).
Five-year CIN2+ risks following non-normal cytology were 22.0% (95%CI=12.0–32.0%) for WLWH and 23.5% (95%CI=21.7–25.3%) for HIV[−] women (p=0.77). Five-year CIN2+ risks for normal cytology were 0.6% (95%CI=0.0–1.1%) for WLWH and 0.8% (95%CI=0.7–0.8%) for HIV[−] women (p=0.50).
Five-year CIN2+ risks for WLWH with positive HPV and non-normal cytology, positive HPV and normal cytology, negative HPV and non-normal cytology, and negative HPV and normal cytology were 24.9% (95%CI=13.4–36.4%), 3.0% (95%CI=0.0–7.4%), 3.6 (95%CI=0.0–9.8%) and 0.3% (95%CI=0.0–0.8%), respectively. Corresponding 5-year CIN2+ risks for HIV[−] women were 26.6% (95%CI=24.6–28.7%), 8.5% (95%CI=7.2–9.9%), 1.9% (95%CI=1.0–2.8%), and 0.5% (95%CI=0.4–0.6%), respectively. Notably, HPV-positive, cytology-negative HIV[−] women had significantly higher 5-year cumulative CIN2+ risk than WLWH (8.5% vs. 3.0%, respectively, p=0.02).
We used a logistic regression and cox proportional hazard model to test whether HIV was an independent risk factor for prevalent and incident CIN2+, respectively, after controlling for HPV and cytology results. Being HIV positive (vs. negative) was neither associated with prevalent (OR=0.8, 95%CI=0.3–1.3) nor incident (HR=0.9, 95%CI=0.3–1.4) CIN2+. Positive (vs. negative) cytology (OR=9.5, 95%CI=3.3–15.7) and positive (vs. negative) HPV (OR=18.4, 95%=6.3–30.6) results were associated with prevalent CIN2+. Likewise, positive (vs. negative) cytology (HR=1.6, 95%CI=1.0–2.1) and positive (vs. negative) HPV (HR=16.4, 95%=11.3–21.5) results were associated with incident CIN2+.
Discussion
WLWH in this cohort were at approximately twice the risk of CIN2+ over a 5-year period compared to HIV[−] women primarily because they were much more likely to have a positive HPV and/or non-normal cytology result. That is, WLWH had greater carriage of HPV than HIV[−] women. Once stratified on (controlling for) screening results, CIN2+ risks were comparable between groups and therefore might be managed similarly, according to the principle of equal management for equal risk (Perkins et al., 2020). Importantly, CIN2+ risks, a proxy for cancer risk, following negative screening results appeared comparable. We suggest that screening intervals following a negative co-test- or HPV test for this population of WLWH, who are likely to have been previously well screened and their HIV infection well managed, might be extended safely. Extending screening intervals will reduce unnecessary screening and care, including treatment of regressive abnormalities, and associated costs. Cervical treatment has been linked to negative reproductive outcomes e.g., preterm delivery (Kyrgiou et al., 2016).
We acknowledge several limitations of this analysis. The relatively small sample size of WLWH required us to use CIN2+ rather than CIN3+ as our primary endpoint. CIN2 is an equivocal high-grade cervical abnormality, likely an admixture CIN3 and HPV infection. Compared with CIN3, CIN2 is poorly reproducible (Carreon et al., 2007; Stoler and Schiffman, 2001), has a distribution of HPV types that in toto is less risky (Castle et al., 2020), and commonly regresses especially in young women (Tainio et al., 2018). Even so, we are underpowered to detect small differences in risk between the two populations. We did not have data on the current and past HIV status e.g., HIV viral load or CD4 counts. Given the high quality of care at KPNC, we assume that most WLWH were receiving standard-of-care therapy for their HIV infection and therefore likely to have low if not undetectable HIV carriage and be in good health.
Finally, despite national and KPNC guidelines on co-testing for cervical-cancer screening of WLWH, these WLWH were screened much more frequently than HIV[−] women. This may have resulted in diagnosing more, regressive CIN2 while possibly censoring some CIN2 that might have been diagnosed eventually as CIN3.
We hypothesize that overscreening of WLWH is a common practice in the US, given the general knowledge of their increased cervical cancer, not accounting for well-screened populations of WLWH with good HIV control. Medicolegal concerns and frequent HIV-care visits of every 6 months may also influence the frequency of cervical-cancer screening in WLWH. However, these and other data (Keller et al., 2015; Keller et al., 2012) support the de-implementation of frequent cervical screening of those WLWH whose cervical cancer risks and HIV infection are likely well managed, especially in WLWH still considering childbearing because of the potentially avoidable reproductive harms (Kyrgiou et al., 2016) that might result from overscreening. Research is needed to assess benefits, harms, and acceptability of extending screening intervals in comparable WLWH.
Funding:
This work was supported by the intramural research program (ZIACP101237-01) of the US National Cancer Institute/NIH/DHHS. Brian Befano was supported by NCI/NIH under Grant T32CA09168.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Declaration of interests
Dr. Castle has received HPV tests and assays for research at a reduced or no cost from Roche, Becton Dickinson, Cepheid, and Arbor Vita Corporation.
References
- 2021. Guidelines for the Prevention and Treatment of Opportunistic Infections in Adults and Adolescents with HIV: Recommendations from the Centers for Disease Control and Prevention,the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America.
- Carreon JD, Sherman ME, Guillén D, Solomon D, Herrero R, Jerónimo J, Wacholder S, Rodríguez AC, Morales J, et al. , 2007. CIN2 is a much less reproducible and less valid diagnosis than CIN3: results from a histological review of population-based cervical samples. Int J Gynecol Pathol 26:441–6. [DOI] [PubMed] [Google Scholar]
- Castle PE, Adcock R, Cuzick J, Wentzensen N, Torrez-Martinez NE, Torres SM, Stoler MH, Ronnett BM, Joste NE, et al. , 2020. Relationships of p16 Immunohistochemistry and Other Biomarkers With Diagnoses of Cervical Abnormalities: Implications for LAST Terminology. Arch Pathol Lab Med 144:725–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Castle PE, Fetterman B, Poitras N, Lorey T, Kinney W, 2012. Safety against cervical precancer and cancer following negative human papillomavirus and Papanicolaou test results in human immunodeficiency virus-infected women. Arch Intern Med 172:1041–3. [DOI] [PubMed] [Google Scholar]
- Cheung LC, Pan Q, Hyun N, Schiffman M, Fetterman B, Castle PE, Lorey T, Katki HA, 2017. Mixture models for undiagnosed prevalent disease and interval-censored incident disease: applications to a cohort assembled from electronic health records. Stat Med 36:3583–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Curry SJ, Krist AH, Owens DK, Barry MJ, Caughey AB, Davidson KW, Doubeni CA, Epling JW Jr., Kemper AR, et al. , 2018. Screening for Cervical Cancer: US Preventive Services Task Force Recommendation Statement. Jama 320:674–86. [DOI] [PubMed] [Google Scholar]
- Gage JC, Schiffman M, Katki HA, Castle PE, Fetterman B, Wentzensen N, Poitras NE, Lorey T, Cheung LC, et al. , 2014. Reassurance against future risk of precancer and cancer conferred by a negative human papillomavirus test. J Natl Cancer Inst 106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grulich AE, van Leeuwen MT, Falster MO, Vajdic CM, 2007. Incidence of cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: a meta-analysis. Lancet 370:59–67. [DOI] [PubMed] [Google Scholar]
- Hyun N, Cheung LC, Pan Q, Schiffman M, Katki HA, 2017. FLEXIBLE RISK PREDICTION MODELS FOR LEFT OR INTERVAL-CENSORED DATA FROM ELECTRONIC HEALTH RECORDS. Ann Appl Stat 11:1063–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Katki HA, Kinney WK, Fetterman B, Lorey T, Poitras NE, Cheung L, Demuth F, Schiffman M, Wacholder S, et al. , 2011. Cervical cancer risk for women undergoing concurrent testing for human papillomavirus and cervical cytology: a population-based study in routine clinical practice. Lancet Oncol 12:663–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Keller MJ, Burk RD, Massad LS, Eltoum IE, Hessol NA, Castle PE, Anastos K, Xie X, Minkoff H, et al. , 2015. Cervical Precancer Risk in HIV-Infected Women Who Test Positive for Oncogenic Human Papillomavirus Despite a Normal Pap Test. Clin Infect Dis 61:1573–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Keller MJ, Burk RD, Xie X, Anastos K, Massad LS, Minkoff H, Xue X, D’Souza G, Watts DH, et al. , 2012. Risk of cervical precancer and cancer among HIV-infected women with normal cervical cytology and no evidence of oncogenic HPV infection. Jama 308:362–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kyrgiou M, Athanasiou A, Paraskevaidi M, Mitra A, Kalliala I, Martin-Hirsch P, Arbyn M, Bennett P, Paraskevaidis E, 2016. Adverse obstetric outcomes after local treatment for cervical preinvasive and early invasive disease according to cone depth: systematic review and meta-analysis. Bmj 354:i3633. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Perkins RB, Guido RS, Castle PE, Chelmow D, Einstein MH, Garcia F, Huh WK, Kim JJ, Moscicki AB, et al. , 2020. 2019 ASCCP Risk-Based Management Consensus Guidelines for Abnormal Cervical Cancer Screening Tests and Cancer Precursors. J Low Genit Tract Dis 24:102–31. [DOI] [PubMed] [Google Scholar]
- Silverberg M, al, e., 2015. Cumulative Incidence of Cancer Among Persons With HIV in North America: A Cohort Study. Ann Intern Med 163:507=18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stelzle D, Tanaka LF, Lee KK, Ibrahim Khalil A, Baussano I, Shah ASV, McAllister DA, Gottlieb SL, Klug SJ, et al. , 2021. Estimates of the global burden of cervical cancer associated with HIV. Lancet Glob Health 9:e161–e69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stoler MH, Schiffman M, 2001. Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL Triage Study. Jama 285:1500–5. [DOI] [PubMed] [Google Scholar]
- Tainio K, Athanasiou A, Tikkinen KAO, Aaltonen R, Cárdenas J, Hernándes, Glazer-Livson S, Jakobsson M, Joronen K, et al. , 2018. Clinical course of untreated cervical intraepithelial neoplasia grade 2 under active surveillance: systematic review and meta-analysis. Bmj 360:k499. [DOI] [PMC free article] [PubMed] [Google Scholar]