A retrospective analysis of bone loss in tenofovir-emtricitabine therapy for HIV PrEP

Abstract

Background

Tenofovir-Emtricitabine (TDF-FTC) is known to cause bone loss in about 1–3% of HIV treated patients. Current studies lack evidence in minority groups and long-term bone loss effects in PrEP patients.

Setting

To address the risk of osteopenia/osteoporosis in patients on TDF-FTC therapy for HIV PrEP and to address the breakthrough incidence of HIV.

Methods

A retrospective analysis was performed in Kaiser Permanente patients from 2012–2021. Patients on TDF-FTC for PrEP without any prior history of osteopenia/osteoporosis (N = 7698) were analyzed to determine the relationship between PrEP adherence and osteopenia/osteoporosis. Descriptive statistics and Cox proportional hazards model were used to compare and analyze patient characteristics between those who developed osteopenia/osteoporosis and those who didn’t.

Results

3% were found to have osteopenia/osteoporosis. Patients who developed osteopenia/osteoporosis were more likely to have a proportion of days covered (PDC) ratio ≥90%, older, had history of Hep B, DM, CVD, CKD, hypertension, and baseline eGFR ≥90 mL/min/1.73 m². Kaplan-Meier curve showed the event-free rate of osteopenia/osteoporosis decreased with time, with a greater reduction in patients with high adherence. Survival analysis showed only PDC of ≥90% was significantly associated with the risk of osteopenia/osteoporosis when adjusted. No incidence of HIV infection was detected.

Conclusions

This retrospective cohort analysis showed that TDF-FTC offered superior PrEP protection. Although high PrEP adherence ensured protection from HIV infection, it was significantly associated with a higher risk of developing osteopenia/osteoporosis. These findings suggest that routine check-ups for osteopenia/osteoporosis may be needed.

Keywords

AIDS HIV prevention antiretroviral therapy screening

Introduction

Tenofovir disoproxil fumarate - emtricitabine (TDF-FTC), is a once-daily prescription medication for adults who are at risk of human immunodeficiency virus (HIV) infection.¹ It’s a nucleoside reverse transcriptase inhibitor, indicated in combination with other antiretroviral agents for the treatment of HIV infection in adults and pediatric patients of 12 years and older. It is also widely prescribed as pre-exposure prophylaxis (PrEP) and helps lower the chance of acquiring HIV through sexual transmission. As PrEP, this medication is prescribed for patients who are HIV negative before and while taking TDF-FTC.¹ Clinical trials showed that PrEP with TDF-FTC reduces the risk of HIV infection in heterosexual men, transgender women, men who have sex with men (MSM), and people who inject drugs.^2–6

Because the treatment of HIV infection is lifelong, research is consistently looking for ways to minimize the side effects of antiretroviral therapies (ARTs). Initiation of different types of ART, including TDF-FTC, have been shown to reduce bone mineral density and increase bone turnover.^7–11 Contrastingly, a large meta-analysis on tenofovir disoproxil fumarate didn’t yield a significant bone loss in PrEP patients.¹² These studies propose conflicting evidence, have small population sizes, and short follow-up length. Long-term large randomized or retrospective studies that address bone loss in FTC-TDF therapy for PrEP are also limited. The goal of this study is to address these limitations by analyzing the risk of osteopenia/osteoporosis in a real life setting for patients on TDF-FTC for PrEP. Specifically, this study aims to examine the relationship between osteopenia/osteoporosis and medication adherence, as well as patient specific factors, such as age, sex, race, liver diseases, kidney diseases, diabetes, cardiovascular diseases, estimated glomerular filtration rate (eGFR), and body mass index (BMI).

Furthermore, TDF-FTC has been shown to lack evidence in Black, Hispanic, and minority groups,⁶ thus a secondary aim of this study is to identify the breakthrough rate of HIV infection among a racial/ethnic diverse cohort of patients on TDF-FTC therapy. The safety findings of this study will help outline the proper precautions and monitoring procedures to minimize adverse events that may arise in patients while on therapy.

Methods

Study design

A retrospective cohort analysis on patients taking TDF-FTC for HIV PrEP was conducted. All patients were members of Kaiser Permanente Southern California (KPSC) and received TDF-FTC for HIV PrEP from January 2012 to December 2020. The relationship of TDF-FTC adherence (defined as proportion of days covered) and incidence of osteopenia or osteoporosis (defined by the World Health Organization as T-score of −1 or lower) were analyzed as the primary endpoint.¹³ The incidence of HIV infection after the initiation of TDF-FTC therapy, and its relation to medication adherence was analyzed as the secondary endpoint. Additional endpoints analyzed were the incidence of osteopenia/osteoporosis based on age, sex, race, hypertension (HTN), cardiovascular disease (CVD), diabetes mellitus (DM), chronic kidney disease (CKD), hepatitis B and C, baseline eGFR, and BMI.

Subjects

Kaiser Permanente Southern California is an integrated health maintenance organization where patients receive health care from a multidisciplinary team. Kaiser Permanente Southern California utilizes a comprehensive electronic medical record system and an integrated pharmacy prescription database. Data from KPSC indicate that members are very similar to the general population of Southern California with regards to age, sex, and race/ethnicity, with only slight underrepresentation of those from lower and higher ends of socioeconomic status.¹⁴

Eligible patients were identified by querying the KPSC database. All prescription data, including sold date and days-supply, were extracted from the pharmacy database. Prescriptions without a valid sold date were excluded. Subjects were HIV-negative patients 18 years of age and older on TDF-FTC therapy at any time from July 2012 to December 2020. A TDF-FTC exposure was defined from the sold date until the end of the days-supply of a TDF-FTC prescription. Multiple exposures were combined into a single TDF-FTC episode if there was no more than 60 days gap between them. The index date was defined as the start of the most recent TDF-FTC episode. Patients were followed until an event of osteopenia/osteoporosis occurred (defined using T-scores from Dual-Energy X-Ray Absorptiometry (DEXA) scans), the end of their treatment (defined as 60 days after the end of their most recent FTC-TDF episode), Kaiser Permanente (KP) membership termination, death, or the end of the study (30 June 2021), whichever date came first. Additional inclusion criteria included KP membership for at least 1 year prior to the index date, allowing a 30-day coverage gap. Patients were excluded if they were HIV positive prior to the index date or pregnant during their follow-up. Additional exclusion criteria were patients taking TDF-FTC as post-exposure prophylaxis, for hepatitis B therapy, or without a PrEP indication. Patients with a history of bone loss before the index date (based on ICD9/10 diagnosis codes recorded during patient encounters and T-scores from DEXA scans), no KP membership, no prescription coverage on the index date, and missing sex identification were excluded.

Out of all the KPSC patients, 31,481 patients were taking TDF-FTC as part of their medication regimen during the study period. However, 18,998 were excluded because they were on TDF-FTC as part of an HIV or post-exposure prophylaxis regimen. From this subject pool, 4785 patients were excluded for the following reasons: <18 years of age, membership less than 1 year before the first TDF-FTC dispense date, HIV positive, prescribed as Hepatitis B or post-exposure prophylaxis therapy, dispensed without a sold date, no KP membership and prescription coverage at the start of the most recent TDF-FTC episode, were pregnant during treatment, missing sex information, history of bone loss before the most recent TDF-FTC episode, or T-score ≤−1.0 before start of therapy (Figure 1).

Figure 1.

Cohort selection criteria flow chart. Legend: Inclusion criteria were HIV-negative patients 18 years of age and older on TDF-FTC therapy at any time from July 2012 to December 2020. A TDF-FTC exposure was defined from the sold date of a TDF-FTC prescription until the end of days-supply of such prescription. Multiple exposures were combined into a single TDF-FTC episode if there was no gap of 60 or more days between them. The index date was defined as the start of the most recent TDF-FTC episode. Additionally, patients must have had KP membership for at least 1 year prior to the index date, allowing a 30 day coverage gap. Exclusion criteria were HIV positive patients prior to the index date or pregnancy during follow-up. Additionally, patients taking TDF-FTC as post-exposure prophylaxis, for hepatitis B therapy, or without a PrEP indication, patients with a history of bone loss before the index date (based on ICD9/10 diagnosis codes and T-scores from DEXA scans), patients without KP membership and prescription coverage on the index date, and patients with missing sex identification were excluded from this study. **denotes number dispensed instead of number of patients.

Measurements

The main outcome was the time from the index date until a patient developed osteopenia or osteoporosis (T-score of −1 or lower), defined using T-scores from DEXA scans during follow-up. T-scores were extracted from the radiology department database, which receives real-time data from every KPSC medical center that operates a Lunar machine. Each DEXA scan produced T-scores from multiple body parts. Patients with either osteopenia or osteoporosis were defined as those who had any T-score ≤−1 during the follow-up period. Z-scores were denotations reduce confusion when looking at exclusion. Some numbers denote dispensed and some denote number of patients utilized for premenopausal women and men <50. Hence, osteoporosis and osteopenia were not defined based on this methodology.

HIV-positive status during follow-up was confirmed by reviewing medical charts, medical center case lists, and the ICD-9/10 diagnosis codes.

Adherence to TDF-FTC was defined using the proportion of days covered (PDC) ratio, a method that has previously been used and validated in administrative pharmacy databases at KP and other institutions.^15,16 PDC is the number of days a patient is covered by the medication divided by the number of days the patient is eligible to have the medication on hand, reported as a percentage. This was calculated using all eligible TDF-FTC prescription fills/refills during the follow-up period. Indefinite stockpiling was allowed by shifting early refills to begin on the day after the end of days-supply of the previous prescription. The PDC ratio was further divided into two categories, PDC ≥90% and PDC <90%.^17,18

Other variables were extracted from the KP HealthConnect electronic medical record system: age, sex, race, HIV status, prior diagnosis of hepatitis B, hepatitis C, diabetes mellitus, cardiovascular disease, chronic kidney disease, and hypertension, baseline eGFR value and BMI category. Age was measured by intervals of 18–29, 30–39, 40–49, or ≥50 years old. Sex was measured on a binary scale of either female or male, assigned at birth. Race and ethnicity were measured as either White, Hispanic, Asian/Pacific Islander, Black, or other/unknown. History of hepatitis B, hepatitis C, DM, CVD, CKD, and HTN identified using ICD-9 and ICD-10 codes and recorded on a yes or no basis. Baseline eGFR was reported as either <70, 70–89, >89 mL/min/1.73 m², or unknown. BMI was reported as underweight (BMI <18.5), normal weight (BMI 18.5–24.9), overweight (BMI 25–29.9), obese (BMI ≥30), or unknown based on definition outlined by the Centers for Disease Control and Prevention.¹⁹ Baseline eGFR and BMI were defined as the most recent value within 12 months before the index date.

Statistical method

Descriptive statistics was used to compare characteristics between patients with and without osteopenia/osteoporosis. Patient characteristics included demographics (age, sex, race/ethnicity), PrEP adherence, baseline eGFR and BMI, history of hepatitis B or C, DM, CVD, CKD, and HTN. Kruskal-Wallis tests were used to compare continuous variables (e.g. PDC) and Chi-squared tests for categorical variables (e.g. patient specific factors, age group, etc.) between the two groups. A Cox proportional hazards model was used to estimate the relationship between patient characteristics and osteopenia/osteoporosis. Kaplan-Meier curve was performed to demonstrate the event-free rate of osteopenia/osteoporosis over time for each adherence group. The log-rank test was used to compare the Kaplan-Meier curves across adherence groups. Following convention, a p-value < .05 was considered statistically significant. Frequency of DEXA scans comparing patients in the ≥90% vs <90% will also be measured.

Approval from the KPSC institutional review board was obtained. The review board waived the requirement for informed consent before the start of the study for all patients due to the retrospective nature of this study.

Results

In this cohort of 7698 patients, the majority was between the ages of 18–39 (69.1%), male (96.9%), Hispanic (40.8%) or White (36.3%). A large percentage of patients had an eGFR value >89 mL/min/1.73 m² (66.2%) and were either normal weight (33.4%) or overweight (34.8%). The incidence of HIV infection while on TDF-FTC therapy was 0% in this cohort.

Table 1 summarizes patient characteristics by their osteopenia/osteoporosis status. Of the 7698 patients, 217 developed osteopenia/osteoporosis (T-score ≤ −1) during the follow-up period. Compared to patients who did not develop osteopenia/osteoporosis, those who did were significantly more likely to have a PDC of ≥90% (90.8% vs. 39.5%, p < .001); were 50 years of age or older (78.8% vs. 11.4%, p < .001); non-Hispanic White (62.2% vs. 35.6%, p < .001); have a history of hepatitis B (2.8% vs. 0.7%, p < .001), diabetes (8.8% vs. 5.4%, p = .035), CVD (12.4% vs. 4.1%, p < .001), CKD (10.6% vs. 5.9%, p = .004), and hypertension (29% vs. 13.1%, p < .001); have a baseline eGFR of less than 90 mL/min/1.73 m² (59% vs. 28.2%, p < .001); and overweight (47.5% vs 34.4%, p < .001).

Table 1.

Summary of patient factors in patients with normal bone mineral density versus osteopenia or osteoporosis.

Category	T-score ≤−1 (any osteoporosis or osteopenia)
Category	No (N = 7481)	Yes (N = 217)	p-value
Length of follow-up (days)^a			0.0013
N	7481	217
Mean (SD)	503.6 (488.53)	468.1 (479.76)
Median	309	343
Q1, Q3	151.0, 683.0	40.0, 834.0
Range	(1.0–2606.0)	(1.0–1932.0)
PDC (%)			<0.0001
N	7481	217
Mean (SD)	78.4 (20.38)	96.5 (8.21)
Median	84.4	99.8
Q1, Q3	63.5, 96.0	96.2, 100.0
Range	(4.7–100.0)	(41.1–100.0)
PDC (%) category			<0.0001
≥90%	2954 (39.5%)	197 (90.8%)
<90%	4527 (60.5%)	20 (9.2%)
Age groups			<0.0001
18–29	2663 (35.6%)	4 (1.8%)
30–39	2646 (35.4%)	7 (3.2%)
40–49	1322 (17.7%)	35 (16.1%)
50+	850 (11.4%)	171 (78.8%)
Sex			0.0642
Female	233 (3.1%)	2 (0.9%)
Male	7248 (96.9%)	215 (99.1%)
Race/ethnicity			<0.0001
White	2662 (35.6%)	135 (62.2%)
Hispanic	3091 (41.3%)	47 (21.7%)
Asian/Pacific Islander	782 (10.5%)	21 (9.7%)
Black	518 (6.9%)	5 (2.3%)
Others or unknown	428 (5.7%)	9 (4.1%)
History of hepatitis B	52 (0.7%)	6 (2.8%)	0.0005
History of hepatitis C	84 (1.1%)	2 (0.9%)	0.781
History of diabetes	407 (5.4%)	19 (8.8%)	0.0352
History of CVD	310 (4.1%)	27 (12.4%)	<0.0001
History of CKD	440 (5.9%)	23 (10.6%)	0.004
History of hypertension	977 (13.1%)	63 (29%)	<0.0001
Baseline eGFR^b			<0.0001
<70 mL/min/1.73 m²	345 (4.6%)	29 (13.4%)
70–89 mL/min/1.73 m²	1763 (23.6%)	99 (45.6%)
>89 mL/min/1.73 m²	5010 (67%)	83 (38.2%)
Unknown	363 (4.9%)	6 (2.8%)
BMI category^b			0.0001
Underweight	76 (1%)	1 (0.5%)
Normal weight	2498 (33.4%)	73 (33.6%)
Overweight	2577 (34.4%)	103 (47.5%)
Obese	1935 (25.9%)	37 (17.1%)
Unknown	395 (5.3%)	3 (1.4%)

Note: p-value was obtained from Kruskal–Wallis test for continuous variables and Chi-squared test for categorical variables.

^aLength of follow-up (in days) from the start of the most recent FTC/TDF episode until the occurrence of an event, end of the FTC/TDF episode (plus 60 days), membership disenrollment, death, or the end of study period (30 June 2021), whichever came first.

^bMost recent value within 12 months before index date.

Figure 2 presents the Kaplan-Meier survival curves for the entire cohort and by PrEP adherence (PDC of 90%–100% and <90%). At some point after 6.5 years, the probability of experiencing osteopenia/osteoporosis was 10.6% for the entire cohort. There was a significant difference in survival times between the two PrEP adherence groups (log-rank test p < .001). For patients with a PDC <90%, their probability of developing osteopenia/osteoporosis at some time after 6.5 years was approximately 3.7%, whereas this estimate was 13.5% for those with a PDC of ≥90%.

Figure 2.

Kaplan-Meier graph, overall and stratified by percentage of days covered (PDC). Legend: Note: p-value was calculated using the log-rank test.

Table 2 presents the unadjusted and adjusted Cox proportional hazard models for the relationship between patient characteristics and the risk of osteopenia/osteoporosis. In unadjusted analyses, the following clinical characteristics (compared to their reference groups) were significantly associated with an increase in risk of osteopenia/osteoporosis: PDC of ≥90% (HR, 8.18; 95% CI, 5.12–13.08), history of hepatitis B (HR, 2.59; 95% CI, 1.15–5.84), CVD (HR, 3.13; 95% CI, 2.09–4.69), CKD (HR, 1.89; 95% CI, 1.22–2.91), hypertension (HR, 2.21; 95% CI, 1.65–2.97), baseline eGFR within 70–89 mL/min/1.73 m² (HR, 2.54; 95% CI, 1.89–3.41), and baseline eGFR <70 mL/min/1.73 m² (HR, 3.29; 95% CI, 2.15–5.04). Contrastingly, obesity was associated with a decline in the risk of osteopenia/osteoporosis (HR, 0.61; 95% CI, 0.41, 0.91). When adjusting for all covariates, only a PDC of ≥90% (HR, 5.36; 95% CI, 3.34–8.61) and obesity (HR, 0.43; 95% CI, 0.29–0.66) had significant impact to the risk of osteopenia/osteoporosis.

Table 2.

Survival analysis of T-score ≤−1 during follow-up for patients with any osteopenia or osteoporosis.

Category	T-score ≤−1 (any osteopenia or osteoporosis)
	N = 7698
	Unadjusted HR [95% CI]	Adjusted HR [95% CI]
PDC (%) category
≥90%	8.18 [5.12, 13.08]	5.36 [3.34, 8.61]
<90%	reference	reference
Age groups
18–29	0.012 [0.005, 0.033]	0.01 [0, 0.03]
30–39	0.017 [0.008, 0.036]	0.02 [0.01, 0.03]
40–49	0.137 [0.095, 0.197]	0.13 [0.09, 0.19]
50+	reference	reference
Sex
Female	0.424 [0.105, 1.706]	0.74 [0.18, 3.09]
Male	reference	reference
Race/ethnicity
White	reference	reference
Asian/Pacific Islander	0.554 [0.35, 0.877]	1.4 [0.87, 2.24]
Black	0.236 [0.097, 0.577]	0.48 [0.19, 1.17]
Hispanic	0.368 [0.264, 0.513]	1.03 [0.74, 1.46]
Others or unknown	0.49 [0.249, 0.962]	0.97 [0.49, 1.92]
History of hepatitis B
No	reference	reference
Yes	2.591 [1.15, 5.839]	1.42 [0.62, 3.26]
History of hepatitis C
No	reference	reference
Yes	0.767 [0.191, 3.086]	0.64 [0.16, 2.61]
History of diabetes
No	reference	reference
Yes	1.537 [0.96, 2.462]	0.85 [0.51, 1.41]
History of CVD
No	reference	reference
Yes	3.134 [2.094, 4.69]	1.21 [0.79, 1.84]
History of CKD
No	reference	reference
Yes	1.885 [1.223, 2.905]	1.15 [0.72, 1.84]
History of hypertension
No	reference	reference
Yes	2.212 [1.648, 2.968]	0.82 [0.6, 1.13]
Baseline eGFR
<70 mL/min/1.73 m²	3.291 [2.149, 5.039]	0.69 [0.45, 1.08]
70-89 mL/min/1.73 m²	2.539 [1.892, 3.406]	0.91 [0.68, 1.23]
>89 mL/min/1.73 m²	reference	reference
Unknown	1.081 [0.472, 2.475]	0.86 [0.37, 2.01]
BMI category (last 12 months)
Underweight	0.607 [0.084, 4.373]	3.95 [0.5, 31.06]
Normal weight	reference	reference
Overweight	1.225 [0.907, 1.654]	0.82 [0.61, 1.12]
Obese	0.614 [0.413, 0.912]	0.43 [0.29, 0.66]
Unknown	0.373 [0.117, 1.184]	0.34 [0.1, 1.08]

HR: hazard ratio; CI: confidence interval.

Note: Adjusted analyses controlled for age, sex, race/ethnicity, history of hepatitis B, hepatitis C, diabetes, CVD, CKD, hypertension, baseline eGFR, and BMI.

Table 3 showed 331 patients in the PDC ≥90% group and 72 patients in the PDC <90% group received DEXA scans. Of these patients, 59.52% in the PDC ≥90% and 27.78% in the PDC <90% cohort have shown a positive event rate (DEXA T-score ≤ −1).

Table 3.

Osteopenia/osteoporosis event frequency among those who got a DEXA scan during the follow-up period.

Any osteopenia/osteoporosis	PDC ≥90%	PDC <90%
Negative event	134 (40.48%)	52 (72.22%)
Positive event	197 (59.52%)	20 (27.78%)
Total	331	72

Discussion

This retrospective cohort study demonstrated a statistically significant increase in the incidence of osteoporosis/osteopenia in HIV negative patients initiated on TDF-FTC for PrEP. This study found a statistically significant decrease in the event-free survival probability for patients with a higher PDC ≥90% versus those with PDC <90%. This pattern persisted after almost 6.5 years of follow-up. High adherence rate and prolonged exposure to TDF-FTC may partially contribute to a higher risk of developing osteopenia/osteoporosis. Our study is significant for its population size, persistence of the incidence of osteoporosis or osteopenia with time, and the ability to adjust for various observable confounding factors and demographics.

An additional finding to our study was that obesity conveyed a protective effect on the incidence of osteoporosis/osteopenia (p = .0001). This contrasts with other studies, which found that obesity has positive association with osteoporosis and osteopenia.^20–22 This study used BMI to define obesity. Other studies have suggested that using BMI as a measure of obesity is controversial because BMI is not directly correlated to the percentage of body fat;²⁰ BMI findings in this study may not be accurate. Other studies have found that percent body fat is proportional to increasing incidence of osteoporosis/osteopenia, with the proposed mechanism being proinflammatory cytokines signaling bone resorption by osteoclasts.²¹ Further research should be conducted analyzing percent body fat versus BMI and osteopenia/osteoporosis.

HIV infection has been indicated as a risk factor for bone loss in children and adults. Osteopenia and osteoporosis are the most important bone disorders affecting patients with HIV infection. Studies have demonstrated that bone mineral density loss is a function of altered bone metabolism, and researchers have hypothesized that this is a result of both the HIV infection and ARTs.^23–26 Studies have shown that tenofovir treatment is associated with inhibited cortical bone mineralization growth in Macaca mulatta monkeys,²⁷ and this has been translated to pre-pubertal children and adolescents infected with HIV.^28,29 Other studies have demonstrated that TDF-FTC has led to decreased bone mineral density, but these studies have limited findings, sample size, and follow up length.³⁰ This study was able to demonstrate that increased TDF-FTC adherence, noted by the PDC ratio, was associated with a significant increase in the incidence of osteoporosis/osteopenia in a large cohort and for up to 6.5 year follow up in PrEP patients.

A proposed mechanism of action for such findings is mitochondrial dysfunction, which induces DNA damage and alters DNA synthesis, resulting in altered gene expressions.^31,32 Another hypothesis is that TDF-FTC is taken up by osteoclasts and osteoblasts, altering gene expression and resulting in increased bone resorption and decreased bone formation.³³ Amongst other hypothesized mechanisms include altered vitamin D metabolism, alternations of calcium homeostasis, and disruptions in parathyroid hormone levels.^34,35 Nonetheless, osteoporosis and osteopenia are multifaceted diseases that develop as a result of patient specific demographics as well as environmental stimuli.^36,37 The answer to tackling the incidence of osteopenia/osteoporosis may not lie in the ART itself, but rather in other approaches. The role of providers should be to undergo continual monitoring³⁸ and treat patient specific risk factors, albeit the benefit and efficacy of TDF-FTC for PrEP outweigh the risk of the development of osteoporosis/osteopenia.

Lastly, this study was able to confirm that TDF-FTC is over 99% effective against HIV infection, as indicated by the CDC.^39–41 In this study, there were no new HIV diagnoses in patients who were HIV negative, as confirmed by reviewing medical charts, medical center case lists, and the ICD-9/10 diagnosis codes. However, this study is limited in that it wasn’t known whether the subjects were IV drug users since it has been noted that IV drug users have decreased protection to 74%.³⁹ Continual monitoring should be practiced in these patients.

The frequency of DEXA scans is higher for patients with higher medication adherence in this study. Some may argue this is a utilization bias since patients who are more consistent in filling their medications are more likely to get regular check-ups and DEXA scans done. However, Kaiser Permanente physicians, as an HMO institution, are in a unique position to order DEXA scans as needed for patients with high risks, e.g. patients over 50 years old or had a bone fracture. Moreover, if patients with high compliance rate translated into higher DEXA scan rate, more cases of osteopenia/osteoporosis might potentially have been detected. Additionally, study patients were followed longitudinally until the event occurred or the end of the study (6 years) per COX analysis, which would have captured patients with a low testing rate. As a result, having more DEXA scan data would not necessarily change the results of our data.

There are multiple limitations to this study. Although DEXA scans are highly accurate, it is possible that cases of osteoporosis or osteopenia went undetected. Additionally, the study is limited by the lack of data on the causes of secondary osteoporosis, such as menopausal status in females, parathyroid hormone diseases, hypogonadism, growth hormone deficiency, and acromegaly.⁴² These causes can pose as potential confounding factors which can affect bone mineral density. With all studies measuring adherence, it is impossible to ensure patients were taking the medication they had picked up. This study also couldn’t explore reasons for TDF-FTC discontinuation or non-adherence. Because this study was an observational study, residual confounding is possible, such as the influence of other medications that could affect bone mineral density. Moreover, this study did not analyze which bones were at most risk of demineralization nor did it analyze the degree of change of bone demineralization between the start of therapy and date of event. Future studies should be conducted in order to guide drug monitoring and safety.

In summary, a higher degree of TDF-FTC adherence was significantly associated with an increase in the risk of osteoporosis/osteopenia in patients initiated on PrEP. This research study is consistent with other studies that demonstrate an increased risk of bone mineral density loss with TDF-FTC therapy.^16,23,24 Although efficacious, the potential long-term effect on bone mineral density may limit the use of TDF-FTC for patients at high risk. Routine check-up for osteopenia may be needed for patients initiated on TDF-FTC for PrEP. Furthermore, could on-demand PrEP be utilized as mainstream therapy in lieu of continuous therapy? With more data supporting the use of on-demand PrEP, this approach may potentially mitigate the risk of serious side effects, such as bone loss.^43,44

Footnotes

Acknowledgements

These data were Presented at Conference on Retroviruses and Opportunistic Infection, Virtual, 14 February 2022. We would like to thank the Kaiser Permanente Regional Bio-Statistical Department for their generous support.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Joseph Chang

Duy Do

References

Truvada [package insert]. Foster City, CA: Gilead Pharmaceuticals Inc., 2016.

McCormack

Dunn

Desai

, et al. Pre-exposure prophylaxis to prevent the acquisition of HIV-1 infection (PROUD): effectiveness results from the pilot phase of a pragmatic open-label randomized trial. Lancet 2016; 387: 53–60.

Grinsztejn

Hoagland

Moreira

, et al. Retention, engagement, and adherence to pre-exposure prophylaxis for men who have sex with men and transgender women in Brasil: 48 week results of a demonstration study. Lancet HIV 2018; 5: e136–e146.

Baeten

Donnell

Ndase

, et al. Antiretroviral prophylaxis for HIV-1 prevention among heterosexual men and women. N Engl J Med 2012; 367(5): 399–410.

Molina

Capitant

Spire

, et al. On-demand preexposure prophylaxis in men at high risk for HIV-1 infection list of authors. N Engl J Med 2015; 373: 2237–2246.

Desai

Field

Grant

, et al. Recent advances in pre-exposure prophylaxis for HIV. BMJ 2017; 359: j5011. DOI: 10.1136/bmj.j5011.

Emtricitabine/Tenofovir Disoproxil Fumarate. Micromedex (accessed October 10, 2021).

Emtricitabine/Tenofovir Disoproxil Fumarate. Lexicomp (accessed October 10, 2021).

Haskelberg

Hoy

Amin

, et al. Changes in bone turnover and bone loss in HIV-infected patients changing treatment to tenofovir-emtricitabine or abacavir-lamivudine. PLoS One 2012; 7(6): e38377. DOI: 10.1371/journal.pone.0038377.

10.

McComsey

Kitch

Daar

, et al. Bone mineral density and fractures in antiretroviral-naive persons randomized to receive abacavir-lamivudine or tenofovir disoproxil fumarate-emtricitabine along with efavirenz or atazanavir-ritonavir: aids Clinical Trials Group A5224s, a substudy of ACTG A5202. J Infect Dis 2011; 203(12): 1791–1801. DOI: 10.1093/infdis/jir188.

11.

Stellbrink

Orkin

Arribas

, et al. Comparison of changes in bone density and turnover with abacavir-lamivudine versus tenofovir-emtricitabine in HIV-infected adults: 48-week results from the ASSERT study. Clin Infect Dis 2010; 51(8): 963–972. DOI: 10.1086/656417.

12.

Baranek

Wang

Cheung

, et al. The effect of tenofovir disoproxil fumarate on bone mineral density: a systematic review and meta-analysis. Antivir Ther 2020; 25(1): 21–32. DOI: 10.3851/IMP3346.

13.

National Institutes of Health . Bone mass measurement: what the numbers mean. October, 2018. Available at: https://www.bones.nih.gov/sites/bones/files/pdfs/bonemassmeasure-508.pdf

14.

Krieger

Overcoming the absence of socioeconomic data in medical records: validation and application of a census-based methodology. Am J Public Health 1992; 82: 703–710.

15.

Bieszk

Patel

Heaberlin

, et al. Detection of medication nonadherence through review of pharmacy claims data. Am J Health Syst Pharm 2003; 60: 360–366.

16.

Horberg

Hurley

Silverberg

, et al. P. Effect of clinical pharmacists on utilization of and clinical response to antiretroviral therapy. J Acquir Immune Defic Syndr 2007; 44: 531–539.

17.

Byrd

Hou

Hazen

, et al. Antiretroviral adherence level necessary for HIV viral suppression using real-world data. J Acquir Immune Defic Syndr 2019; 82(3): 245–251. DOI: 10.1097/QAI.0000000000002142.

18.

Byrd

Hou

Bush

, et al. Adherence and viral suppression among participants of the patient-centered Human Immunodeficiency Virus (HIV) care model project: a collaboration between community-based pharmacists and HIV clinical providers. Clin Infect Dis 2020; 70(5): 789–797. DOI: 10.1093/cid/ciz276.

19.

Centers for Disease Control and Prevention . Defining adult overweight & obesity. June 7, 2021. Available at: https://www.cdc.gov/obesity/adult/defining.html

20.

Migliaccio

Greco

Fornari

, et al.

Is obesity in women protective against osteoporosis?

Diabetes Metab Syndr Obes 2011; 4: 273–282. DOI: 10.2147/DMSO.S11920.

21.

Cao

. Effects of obesity on bone metabolism. J Orthop Surg Res 2011; 6: 30. DOI: 10.1186/1749-799X-6-30.

22.

Tang

Xiao

Lei

, et al. A bivariate whole-genome linkage scan suggests several shared genomic regions for obesity and osteoporosis. J Clin Endocrinol Metab 2007; 92(7): 2751–2757. DOI: 10.1210/jc.2006-2607.

23.

Tebas

Powderly

Claxton

, et al. Accelerated bone mineral loss in HIV-infected patients receiving potent antiretroviral therapy. AIDS 2000; 14(4): F63–F67. DOI: 10.1097/00002030-200003100-00005.

24.

Viganò

Mora

. Adverse effects of antiretroviral therapy: focus on bone density. Expert Opin Drug Saf 2004; 3(3): 199–208. DOI: 10.1517/14740338.3.3.199.

25.

Madeddu

Spanu

Solinas

, et al. Bone mass loss and vitamin D metabolism impairment in HIV patients receiving highly active antiretroviral therapy. Q J Nucl Med Mol Imaging 2004; 48(1): 39–48.

26.

Arpadi

Horlick

Thornton

, et al. Bone mineral content is lower in prepubertal HIV-infected children. J Acquir Immune Defic Syndr 2002; 29(5): 450–454. DOI: 10.1097/00126334-200204150-00004.

27.

Castillo

Tarantal

Watnik

, et al. Tenofovir treatment at 30 mg/kg/day can inhibit cortical bone mineralization in growing rhesus monkeys (Macaca mulatta). J Orthop Res 2002; 20(6): 1185–1189. DOI: 10.1016/S0736-0266(02)00074-8.

28.

Purdy

Gafni

Reynolds

, et al. Decreased bone mineral density with off-label use of tenofovir in children and adolescents infected with human immunodeficiency virus. J Pediatr 2008; 152(4): 582–584. DOI: 10.1016/j.jpeds.2007.12.020.

29.

Hazra

Gafni

Maldarelli

, et al. Tenofovir disoproxil fumarate and an optimized background regimen of antiretroviral agents as salvage therapy for pediatric HIV infection. Pediatrics 2005; 116(6): e846–e854. DOI: 10.1542/peds.2005-0975.

30.

Spinelli

Glidden

Anderson

, et al. Impact of Estimated Pre-Exposure Prophylaxis (PrEP) adherence patterns on bone mineral density in a large PrEP demonstration project. AIDS Res Hum Retroviruses 2019; 35(9): 788–793. DOI: 10.1089/AID.2018.0297.

31.

Kohler

Hosseini

Hoying-Brandt

, et al. Tenofovir renal toxicity targets mitochondria of renal proximal tubules. Lab Invest 2009; 89(5): 513–519. DOI: 10.1038/labinvest.2009.14.

32.

Vidal

Domingo

Guallar

, et al. In vitro cytotoxicity and mitochondrial toxicity of tenofovir alone and in combination with other antiretrovirals in human renal proximal tubule cells. Antimicrob Agents Chemother 2006; 50(11): 3824–3832. DOI: 10.1128/AAC.00437-06.

33.

Grigsby

Pham

Mansky

, et al. Tenofovir-associated bone density loss. Ther Clin Risk Manag 2010; 6: 41–47.

34.

Thomas

Doherty

. HIV infection--a risk factor for osteoporosis. J Acquir Immune Defic Syndr 2003; 33(3): 281–291. DOI: 10.1097/00126334-200307010-00001.

35.

Mondy

Powderly

Claxton

, et al. Alendronate, vitamin D, and calcium for the treatment of osteopenia/osteoporosis associated with HIV infection. J Acquir Immune Defic Syndr 2005; 38(4): 426–431. DOI: 10.1097/01.qai.0000145352.04440.1e.

36.

Kanis

McCloskey

. Risk factors in osteoporosis. Maturitas 1998; 30(3): 229–233. DOI: 10.1016/s0378-5122(98)00090-5.

37.

Lane

. Epidemiology, etiology, and diagnosis of osteoporosis. Am J Obstet Gynecol 2006; 194(2 Suppl): S3–S11. DOI: 10.1016/j.ajog.2005.08.047.

38.

Feldstein

Elmer

Smith

, et al. Electronic medical record reminder improves osteoporosis management after a fracture: a randomized, controlled trial. J Am Geriatr Soc 2006; 54(3): 450–457. DOI: 10.1111/j.1532-5415.2005.00618.x.

39.

Centers for Disease Control and Prevention . Preexposure Prophylaxis for the Prevention of HIV Infection in the United States. U.S. Public Health Services, 2021. Available at: https://www.cdc.gov/hiv/pdf/risk/prep/cdc-hiv-prep-guidelines-2021.pdf

40.

Donnell

Baeten

Bumpus

, et al. HIV protective efficacy and correlates of tenofovir blood concentrations in a clinical trial of PrEP for HIV prevention. J Acquir Immune Defic Syndr 2014; 66(3): 340–348. DOI: 10.1097/QAI.0000000000000172.

41.

Thomson

Baeten

Mugo

, et al. Tenofovir-based oral preexposure prophylaxis prevents HIV infection among women. Curr Opin HIV AIDS 2016; 11(1): 18–26. DOI: 10.1097/COH.0000000000000207.

42.

Ganesan

Jandu

Roane

. . Secondary Osteoporosis. StatPearls. 2022. Available at: https://www.ncbi.nlm.nih.gov/books/NBK470166/ (accessed 7 July 2022).

43.

Stansfield

Moore

Boily

, et al. Estimating benefits from using on-demand oral PrEP by MSM in US: a modeling study. In: Presented at Conference on Retroviruses and Opportunistic Infections, Virtual Conference, 14 February 2022. Available at: https://www.croiconference.org/search-abstracts/.

44.

Antoni

Tremblay

Delaugerre

, et al. On-demand pre-exposure prophylaxis with tenofovir disoproxil fumarate plus emtricitabine among men who have sex with men with less frequent sexual intercourse: a post-hoc analysis of the ANRS IPERGAY trial. Lancet HIV 2020; 7(2): e113–e120. DOI: 10.1016/S2352-3018(19)30341-8.