ABSTRACT
A substantial proportion of people with HIV (PWH) fail to achieve full immune recovery despite long-term antiretroviral therapy (ART), potentially increasing their risk of serious comorbidities. This study investigated the association between immune non-responder (INR) and the incidence of AIDS-defining diseases (ADs) and non-AIDS-defining diseases (NADs) in a prospective cohort at the Third People's Hospital of Shenzhen, China. The low – and high-threshold cohorts included 7,874 and 8,077 individuals with baseline CD4+ T-cell counts < 350 and < 500cells/μL, respectively. INR was defined as failure to reach CD4+ T-cell thresholds (350 or 500 cells/μL) in two consecutive measurements during follow-up. Kaplan-Meier curves and Cox proportional hazards models were used to assess associations. Median follow-up after immune classification was 49.4 and 42.2 months in the low – and high-threshold cohorts, respectively. In the low-threshold cohort, INR was independently associated with significantly increased risks of ADs, including pneumocystis pneumonia (adjusted hazard ratio [aHR], 10.10; 95% confidence interval [CI]: 4.94–20.70), talaromycosis marneffei (aHR, 7.38; 95% CI: 3.51–15.50), and AIDs-defining cancers (aHR, 3.67; 95% CI: 1.20–11.20); and NADs, including end-stage liver disease (aHR, 15.00; 95% CI: 5.59–40.00), cardiovascular disease (aHR, 3.83; 95% CI: 2.14–6.87), chronic kidney disease (aHR, 1.78; 95% CI: 1.23–2.58), and non-AIDS-defining cancers (aHR, 4.75; 95% CI: 2.31–9.74). Similar associations were observed in the high-threshold cohort. INR is strongly associated with long-term morbidity in PWH. These findings highlight the need for improved risk assessment beyond CD4+ T-cell monitoring to reduce disease burden in PWH.
KEYWORDS: Immune non-responder, AIDS-defining diseases, non-AIDS-defining diseases, CD4+ T-cell, morbidity
Introduction
Although antiretroviral therapy (ART) effectively suppresses viral replication, resulting in increased CD4+ T-cell counts and prolonged survival, approximately 10-40% of people with HIV (PWH) fail to achieve full CD4+ T-cell reconstitution [1,2]. These individuals referred to as immune non-responders (INR), experience persistent immune dysfunction, placing them at higher risk for adverse health outcomes.
The mechanisms underlying INR are multifactorial, involving impaired bone marrow and thymic function, persistent immune activation and inflammation, as well as contributing factors such as chronic comorbidities, co-infections, and individual genetic factors [3]. This long-term immune dysregulation is extensively documented as being associated with an increased risk of AIDS-defining diseases (ADs), such as tuberculosis (TB), Kaposi sarcoma, and pneumocystis pneumonia (PCP) [4-6]. For instance, the incidence of Kaposi sarcoma is significantly higher in INR individuals compared to those with complete immune reconstitution (16.91% vs 0.89%) [5], and similar trend was observed for PCP (56.8% vs 21.1%) [6].
Despite the significant improvements in HIV treatment, the incidence of non-AIDS-defining diseases (NADs) continues to rise within this population [7-9]. Studies indicate that the comorbidity-free lifespan of PWH is notably shorter compared to that of the general population [10]. This disparity arises from a combination of factors, including the direct effects of chronic HIV infection, the long-term side effects of ART, and persistent immune dysfunction observed in INR. However, unlike ADs, which have been extensively studied in the context of INR, much less is known about how INR influences NADs risk over time, particularly after reaching widely accepted immune recovery thresholds, such as CD4+ T-cell counts of 350 or 500 cells/μL [11]. As early treatment initiation and ART strategies continue to evolve, understanding the relationship between INR and NADs is crucial for optimizing long-term management strategies for PWH and reassessing the definition of complete immune reconstitution.
In light of these considerations, this study aims to investigate the long-term impact of INR on the incidence of both ADs and NADs in a cohort of PWH. INR was defined as failure to achieve CD4+ T-cell counts of 350 or 500 cells/μL after four or five years of ART, respectively, based on two consecutive measurements. Participants were subsequently followed to assess morbidity outcomes.
Methods
Study design and participants
The study protocol adhered to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Institutional Review Board of Shenzhen Third People’s Hospital (No. 2022-143). All participants provided written informed consent.
We conducted a prospective cohort study of treatment-naive PWH who sought care at the Third People’s Hospital of Shenzhen, China, between 2010 and 2018. This hospital serves as the sole designated facility for HIV treatment and management in Shenzhen.
Two cohorts were constructed based on distinct definitions of INR, each determined by different follow-up durations and CD4+ T-cell recovery thresholds: the low-threshold cohort and the high-threshold cohort. The rationale for the two-threshold definition of INR is detailed in the Supplementary Method. In the low-threshold cohort, INR was defined as failure to achieve CD4+ T-cell count ≥ 350 cells/µL in two consecutive follow-up visits within four years of ART initiation. Inclusion criteria included ART initiation between January 1, 2010, and December 31, 2018, a baseline CD4+ T-cell count < 350 cells/µL, and a minimum follow-up duration of four years and three months. Individuals who developed outcomes (e.g. ADs and NADs) during the first four years were excluded from the analysis of those specific outcomes but remained eligible for others. In the high-threshold cohort, INR was defined as failure to achieve CD4+ T-cell count ≥ 500 cells/µL in two consecutive follow-up visits within five years. This cohort included individuals who initiated ART between January 1, 2010, and December 31, 2017, with a baseline CD4+ T-cell count <500 cells/µL and at least five years and three months of follow-up. Participants who developed outcomes during the first five years were similarly excluded from the analysis of those outcomes but retained for others. To assess the potential for selection bias introduced by exclusion criteria, we compared baseline characteristics between excluded and included individuals for each outcome across both the low – and high-threshold cohorts.
Follow-up and outcomes
We assessed a range of clinical outcomes including both ADs and NADs. ADs included PCP, TB, talaromycosis marneffei (TSM), and AIDS-defining cancers (ADC), as defined by the 2024 edition of the Chinese Guidelines for the Diagnosis and Treatment of HIV/AIDS [12]. NADs encompassed cardiovascular disease (CVD), end-stage liver disease (ESLD), chronic kidney disease (CKD), and non-AIDS-defining cancers (NADC). ESLD was defined as liver failure, compensated or decompensated liver cirrhosis, ascites, spontaneous bacterial peritonitis, variceal hemorrhage, hepatic encephalopathy, or hepatocellular carcinoma [13]. The CVD composite included myocardial infarction, acute coronary syndrome, stroke, transient ischemic attack, and peripheral artery disease [14]. CKD was diagnosed according to KDIGO guidelines as two consecutive estimated glomerular filtration rates (eGFR) measurements < 60 mL/min/1.73 m2, obtained at least three months apart [15]. NADC was defined as any malignancy not classified as AIDS-related, excluding precancerous conditions, primary central nervous system lymphoma, Burkitt lymphoma, immunoblastic lymphoma, Kaposi's sarcoma, non-Hodgkin lymphoma, and cervical cancer [16]. All clinical outcomes were identified using ICD-10 codes and confirmed through hospital electronic medical records system, relevant laboratory, imaging and pathology findings, in accordance with standardized diagnostic guidelines. To reduce the risk of outcome misclassification, each diagnosis was independently reviewed by two experienced HIV clinicians. Any discrepancies were adjudicated by a third senior physician through consensus (Supplementary Table 1).
The baseline for the low-threshold cohort was defined as the date marking four years after ART initiation. Similarly, the baseline for the high-threshold cohort was set at five years post-ART initiation. Outcomes were assessed from three months after cohort entry until the occurrence of death, loss to follow-up, or the end of the follow-up on Sep 30, 2024, whichever came first. Participants were considered lost to follow-up if they missed scheduled visits for more than three consecutive months.
Assessment of covariates
Data were collected by trained interviewers through the hospital’s routine diagnostic and treatment information system. Covariates included the following domains: (1) Demographic characteristics: age, sex, marital status (married or cohabiting, divorced or separated, widowed, never married), body mass index (BMI), smoking status (ever vs. never), and alcohol consumption (yes or no); (2) Laboratory parameters: fasting glucose, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), total cholesterol (TC), creatinine (Cr), eGFR, alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), gamma-glutamyltransferase (GGT), white blood cells (WBC) count, and platelet count; (3) Comorbidities: hepatitis B virus (HBV) infection (defined by a positive HBV surface antigen positive envelope antigen, or detectable HBV DNA) and hepatitis C virus (HCV) infection (defined by a positive anti-HCV antibody or detectable HCV RNA); (4) HIV-related factors: route of HIV transmission, CD8+ T-cell count, HIV RNA levels, presence of opportunistic infection (yes or no), WHO clinical stage, and the time interval between HIV diagnosis and ART initiation.
Statistical analysis
All statistical analyses were performed using R software (version 4.4.0; https://www.r-project.org). Continuous variables were presented as medians with interquartile ranges (IQR), while categorical variables were summarized as frequencies and percentages. Group comparisons for continuous variables were performed using the Kruskal – Wallis test, and categorical variables were compared using the chi-square test.
Cumulative incidence of NADs and ADs incidence using the Kaplan – Meier method, and compared the groups using the log-rank test. We employed Cox proportional hazards model to calculate the hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) for factors associated with each outcomes. HRs were estimated using both univariate and multivariable models, adjusting for demographics, laboratory parameters, comorbidities, and HIV-related factors. Statistical significance was defined as a P-value of less than 0.05.
Since our data were derived from hospital electronic medical records, not all covariate data were complete. For variables with a missing rate <1%, we applied multiple imputation using a random forest algorithm to minimize potential bias and preserve statistical power.
Sensitivity analysis
We conducted a series of sensitivity analyses to test the robustness of our findings and address potential biases. First, a directed acyclic graph (DAG) was constructed based on prior literature and expert consultation to distinguish confounders from potential mediators. The construction followed the evidence synthesis for constructing directed acyclic graphs protocol, integrating causal inference principles with systematic evidence mapping. Guided by the DAG, we excluded variables identified as mediators and re-estimated the multivariable Cox models to assess the stability of effect estimates. Second, to reduce potential bias from short-term follow-up or early events, we excluded individuals followed for less than six months, regardless of whether they experienced outcomes. Third, to address potential confounding arising from baseline imbalances between INR and IR groups, we performed an inverse probability of treatment weighting (IPTW) analysis. Propensity scores were estimated using logistic regression based on key baseline covariates. Covariate balance before and after weighting was assessed using standardized mean differences (SMDs), with a SMD < 0.1 indicating adequate balance. To further evaluate the robustness of the IPTW-adjusted hazard ratios against unmeasured confounding, we conducted quantitative bias analysis using E-values. Finally, to account for the influence of competing event, we used Fine – Gray sub-distribution hazard models, treating death and loss to follow-up as competing events for each clinical outcome. Sub-distribution hazard ratios (sHRs) with 95% CIs were estimated, and cumulative incidence function (CIF) plots were generated to visualize time-to-event distributions. All competing risk analyses were conducted separately within the low – and high-threshold cohorts.
Results
Baseline characteristics of participants
From Jan 1, 2010 to Dec 31, 2018, a total of 8,262 PWH with baseline CD4+ T-cell counts <350 cells/µL initiated ART. After excluding 220 individuals with missing baseline data and 168 without follow-up information, 7,874 participants (95.30%) were included in the analysis to assess immune response at the fourth year of ART and its association with subsequent ADs and NADs. For the high-threshold cohort, 8,467 PWH initiated HIV care at the same hospital between Jan 1, 2010 and Dec 31, 2017. After excluding 390 individuals lacking baseline or follow-up data, 8,077 participants (95.39%) were included in the analysis beginning from the fifth year after ART initiation (Figure 1). Comparisons of baseline characteristics between excluded and included individuals showed no significant differences for the majority of variables (P > 0.05; Supplementary Tables 2–9).
Figure 1.
Study flow diagram. Abbreviations: PWH, people with HIV; ART, antiretroviral therapy; INR, immune non-responders; IR, immune responders; ADs, AIDS-defining diseases; NADs, non-AIDS-defining diseases; PCP, pneumocystis pneumonia; TB, tuberculosis; TSM, talaromycosis marneffei; ADC, AIDS-defining cancers; ESLD, end-stage liver disease; CVD, cardiovascular disease; CKD, chronic kidney disease; NADC, non-AIDS-defining cancers.
At ART initiation in the low-threshold cohort, INR individuals were more likely to be female, married or cohabiting, underweight, and current or former smokers compared to IRs. They also had higher rates of advanced WHO clinical stage, and opportunistic infections (all P values < 0.05). By the fourth year of follow-up, INR individuals more often presented with HCV co-infection, CD8+ T-cell count < 500 cells/μL, and unsuppressed HIV RNA were more frequently observed among INR individuals (all P < 0.05). Additionally, compared to the IRs, the INR group exhibited older age, and lower levels of lipids (e.g. TC, LDL, TG), liver function (e.g. ALT, GGT), and kidney function markers (e.g. Cr, eGFR) (Table 1).
Table 1.
Baseline characteristics of participants at the fourth and fifth years.
| Characteristicsa | Low-threshold cohort | High-threshold cohort | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Overall | INR (n, %) | IR (n, %) | P Value | Overall | INR (n, %) | IR (n, %) | P Value | ||
| Total, n | 7874 | 1891 | 5983 | 8077 | 3228 | 4849 | |||
| Sex | 0.007 | <0.001 | |||||||
| Male | 7072 (89.8) | 1667 (88.2) | 5405 (90.3) | 7276 (90.1) | 2854 (88.4) | 4422 (91.2) | |||
| Female | 802 (10.2) | 224 (11.8) | 578 (9.7) | 801 (9.9) | 374 (11.6) | 427 (8.8) | |||
| Age, years | 36.00 [30.00, 44.00] | 38.00 [31.00, 48.00] | 35.00 [30.00, 43.00] | <0.001 | 36.00 [31.00, 44.00] | 38.00 [32.00, 47.00] | 35.00 [31.00, 42.00] | <0.001 | |
| Marital status* | <0.001 | <0.001 | |||||||
| Never married | 4412 (56.0) | 915 (48.4) | 3497 (58.4) | 4601 (57.0) | 1629 (50.5) | 2972 (61.3) | |||
| Married or cohabiting | 2720 (34.5) | 751 (39.7) | 1969 (32.9) | 2741 (33.9) | 1264 (39.2) | 1477 (30.5) | |||
| Divorced, separated, or widowed | 742 (9.4) | 225 (11.9) | 517 (8.6) | 735 (9.1) | 335 (10.4) | 400 (8.2) | |||
| BMI, kg/m2* | <0.001 | <0.001 | |||||||
| <18.5 | 1381 (17.5) | 406 (21.5) | 975 (16.3) | 1343 (16.6) | 679 (21.0) | 664 (13.7) | |||
| 18.5-23.9 | 5274 (67.0) | 1233 (65.2) | 4041 (67.5) | 5420 (67.1) | 2137 (66.2) | 3283 (67.7) | |||
| ≥24 | 1219 (15.5) | 252 (13.3) | 967 (16.2) | 1314 (16.3) | 412 (12.8) | 902 (18.6) | |||
| Smoking* | 1786 (22.7) | 480 (25.4) | 1306 (21.8) | 0.001 | 1927 (23.9) | 800 (24.8) | 1127 (23.2) | 0.118 | |
| Drinking* | 2007 (25.5) | 462 (24.4) | 1545 (25.8) | 0.238 | 2162 (26.8) | 808 (25.0) | 1354 (27.9) | 0.004 | |
| HBV infection | 879 (12.2) | 227 (13.5) | 652 (11.9) | 0.079 | 884 (11.9) | 378 (12.9) | 506 (11.3) | 0.039 | |
| HCV infection | 131 (1.8) | 45 (2.7) | 86 (1.6) | 0.004 | 137 (1.8) | 76 (2.6) | 61 (1.4) | <0.001 | |
| HIV transmission route | <0.001 | <0.001 | |||||||
| Male-to-male sex contact | 4913 (62.4) | 1026 (54.3) | 3887 (65.0) | 5158 (63.9) | 1876 (58.1) | 3282 (67.7) | |||
| Heterosexual contact | 2720 (34.5) | 756 (40.0) | 1964 (32.8) | 2690 (33.3) | 1216 (37.7) | 1474 (30.4) | |||
| IDU | 74 (0.9) | 41 (2.2) | 33 (0.6) | 77 (1.0) | 53 (1.6) | 24 (0.5) | |||
| Other | 167 (2.1) | 68 (3.6) | 99 (1.7) | 152 (1.9) | 83 (2.6) | 69 (1.4) | |||
| CD8 count, cells/μL | <0.001 | <0.001 | |||||||
| <500 | 1384 (17.6) | 452 (23.9) | 932 (15.6) | 1414 (17.5) | 760 (23.5) | 654 (13.5) | |||
| 500–999 | 4550 (57.8) | 1017 (53.8) | 3533 (59.1) | 4779 (59.2) | 1831 (56.7) | 2948 (60.8) | |||
| ≥1000 | 1940 (24.6) | 422 (22.3) | 1518 (25.4) | 1884 (23.3) | 637 (19.7) | 1247 (25.7) | |||
| HIVRNA, copies/mL | 0.008 | <0.001 | |||||||
| Below detection limit | 7704 (97.8) | 1835 (97.0) | 5869 (98.1) | 7691 (95.2) | 2887 (89.4) | 4804 (99.1) | |||
| Above detection limit | 170 (2.2) | 56 (3.0) | 114 (1.9) | 386 (4.8) | 341 (10.6) | 45 (0.9) | |||
| Opportunistic Infections* | 528 (6.7) | 167 (8.8) | 361 (6.0) | <0.001 | 511 (6.3) | 291 (9.0) | 220 (4.5) | <0.001 | |
| WHO Clinical Stage* | <0.001 | <0.001 | |||||||
| I | 109 (1.4) | 21 (1.1) | 88 (1.5) | 1421 (17.6) | 268 (8.3) | 1153 (23.8) | |||
| II | 3938 (50.0) | 608 (32.2) | 3330 (55.7) | 3468 (42.9) | 1007 (31.2) | 2461 (50.8) | |||
| III | 2110 (26.8) | 626 (33.1) | 1484 (24.8) | 1678 (20.8) | 1020 (31.6) | 658 (13.6) | |||
| IV | 1717 (21.8) | 636 (33.6) | 1081 (18.1) | 1510 (18.7) | 933 (28.9) | 577 (11.9) | |||
| Time interval, month | 0.004 | <0.001 | |||||||
| <1 | 3880 (49.3) | 995 (52.6) | 2885 (48.2) | 3623 (44.9) | 1582 (49.0) | 2041 (42.1) | |||
| 1–5 | 2282 (29.0) | 515 (27.2) | 1767 (29.5) | 2386 (29.5) | 945 (29.3) | 1441 (29.7) | |||
| ≥6 | 1712 (21.7) | 381 (20.1) | 1331 (22.2) | 2068 (25.6) | 701 (21.7) | 1367 (28.2) | |||
| ART regimen | <0.001 | <0.001 | |||||||
| 2NRTIs + EFV | 6490 (82.4) | 1449 (76.6) | 5041 (84.3) | 6835 (84.6) | 2580 (79.9) | 4255 (87.8) | |||
| 2NRTIs + LPV/r | 458 (5.8) | 157 (8.3) | 301 (5.0) | 432 (5.3) | 230 (7.1) | 202 (4.2) | |||
| 2NRTIS + NVP | 466 (5.9) | 102 (5.4) | 364 (6.1) | 491 (6.1) | 227 (7.0) | 264 (5.4) | |||
| DTG-containing | 138 (1.8) | 53 (2.8) | 85 (1.4) | 66 (0.8) | 40 (1.2) | 26 (0.5) | |||
| Others | 322 (4.1) | 130 (6.9) | 192 (3.2) | 253 (3.1) | 151 (4.7) | 102 (2.1) | |||
| Glucose, mmol/L | 5.10 [4.78, 5.47] | 5.11 [4.76, 5.53] | 5.10 [4.79, 5.46] | 0.59 | 5.10 [4.77, 5.47] | 5.11 [4.77, 5.51] | 5.09 [4.76, 5.44] | 0.013 | |
| WBC, 10^9/L | 5.90 [4.99, 7.00] | 5.52 [4.61, 6.58] | 6.02 [5.10, 7.09] | <0.001 | 6.00 [5.06, 7.14] | 5.57 [4.69, 6.63] | 6.28 [5.34, 7.41] | <0.001 | |
| Platelet, 10^9/L | 234.00 [200.00, 273.00] | 225.00 [186.00, 265.00] | 237.00 [204.00, 275.00] | <0.001 | 237.00 [202.00, 276.00] | 227.00 [190.00, 267.00] | 244.00 [210.00, 280.00] | <0.001 | |
| Low-density lipoprotein cholesterol, mmol/L | 2.63 [2.20, 3.12] | 2.52 [2.12, 3.04] | 2.65 [2.23, 3.14] | <0.001 | 2.64 [2.21, 3.1] | 2.54 [2.13, 3.03] | 2.70 [2.27, 3.19] | <0.001 | |
| High-density lipoprotein cholesterol, mmol/L | 1.28 [1.08, 1.50] | 1.28 [1.07, 1.51] | 1.28 [1.09, 1.50] | 0.647 | 1.26 [1.07, 1.47] | 1.28 [1.09, 1.51] | 1.24 [1.06, 1.45] | <0.001 | |
| Triglycerides, mmol/L | 1.40 [0.94, 2.17] | 1.36 [0.92, 2.09] | 1.42 [0.95, 2.20] | 0.015 | 1.43 [0.97, 2.23] | 1.32 [0.90, 2.06] | 1.51 [1.01, 2.35] | <0.001 | |
| Total cholesterol, mmol/L | 4.51 [3.94, 5.18] | 4.40 [3.81, 5.11] | 4.54 [3.98, 5.21] | <0.001 | 4.50 [3.93, 5.16] | 4.38 [3.84, 5.05] | 4.57 [4.00, 5.23] | <0.001 | |
| Creatinine, μmol/L | 73.00 [65.00, 82.00] | 72.00 [63.00, 81.00] | 73.00 [65.00, 82.00] | 0.001 | 73.00 [65.00, 83.00] | 72.00 [64.00, 82.00] | 74.00 [66.00, 83.00] | <0.001 | |
| eGFR, mL/min/ 1.73 m2 | 117.72 [107.18, 124.50] | 116.03 [104.36, 123.98] | 117.99 [107.98, 124.58] | <0.001 | 117.35 [106.50, 124.50] | 116.50 [104.64, 124.10] | 117.96 [107.69, 124.75] | <0.001 | |
| Alanine aminotransferase, U/L | 26.00 [17.00, 38.00] | 23.00 [16.00, 34.00] | 26.00 [18.00, 39.00] | <0.001 | 26.00 [18.00, 38.00] | 24.00 [17.00, 35.00] | 27.00 [19.00, 40.00] | <0.001 | |
| Aspartate aminotransferase, U/L | 23.00 [19.00, 29.37] | 24.00 [19.00, 30.00] | 23.00 [19.00, 29.00] | 0.549 | 23.00 [19.00, 30.00] | 23.00 [19.00, 29.00] | 23.00 [19.00, 30.00] | 0.051 | |
| Total bilirubin, μmol/L | 8.70 [6.50, 11.60] | 8.70 [6.50, 11.90] | 8.60 [6.50, 11.50] | 0.148 | 8.40 [6.40, 11.40] | 8.60 [6.50, 11.80] | 8.20 [6.30, 11.20] | <0.001 | |
| Gamma-glutamyltransferase, U/L | 39.00 [27.00, 62.00] | 37.00 [26.00, 58.00] | 39.00 [27.00, 63.00] | <0.001 | 40.00 [27.00, 62.00] | 38.00 [26.00, 58.00] | 41.00 [28.00, 66.00] | <0.001 | |
Note: a Continuous variables are expressed as median (IQR). Categorical variables are expressed as frequency (percentage).
* The data for these variables were collected at the start of initial treatment.
INR: Immunological non-responder; IR: Immunological responder; BMI: body-mass index; IDU: injection drug use; WBC: white blood cells; eGFR: estimated glomerular filtration rate; Time interval: the time between the diagnosis of HIV and the initiation of ART; HBV: hepatitis B virus; HCV: hepatitis C virus; HIV: human immunodeficiency virus; ART: antiretroviral therapy;3TC: lamivudine; TDF: tenofovir disoproxil fumarate; EFV: efavirenz; NVP: nevirapine; DTG: Dolutegravir; AZT: zidovudine; LPVr: lopinavir/ritonavir; D4T: stavudine; EVG: Elvitegravir; FTC: Emtricitabine; TAF: Tenofovir Alafenamide.
Similar patterns were observed in the high-threshold cohort. In addition, HBV co-infection was more frequently associated with INR status. Compared to the IR group, INR individuals exhibited higher levels of glucose, HDL-C, and TBIL, although these differences were not statistically significant in the low-threshold cohort (Table 1).
Comparison of cumulative incidence of ADs between INR and IR PWH
To further elucidate the immunological context of ADs, we analysed longitudinal CD4+ T-cell dynamics in individuals who developed ADs during follow-up. As shown in Supplementary Figure 1, median CD4+ T-cell counts generally increased over time in both cohorts, with many individuals achieving levels above 350 or 500 cells/μL. Additionally, analysis of CD4+ T-cell levels at the time of AD diagnosis revealed that in the low-threshold cohort, the median CD4+ T-cell count was 267 cells/μL (IQR: 168–370), and 26.91% of ADs occurred in individuals with CD4+ T-cell count ≥350 cells/μL. In the high-threshold cohort, the median count was 288 cells/μL (IQR: 203–419.25), with 14.5% of ADs occurring at CD4+ T-cell count ≥500 cells/μL (Supplementary Table 10).
In the low-threshold cohort, the median follow-up for the four ADs was 49.40–49.45 months (range). During these follow-up periods, 39 participants developed PCP, 64 developed TB, 35 developed TSM, and 18 developed ADC. The 5-year cumulative incidence rates of PCP, TB, TSM, and ADC in the INR group were 32.44‰, 22.39‰, 28.14‰, and 8.73‰, respectively, all significantly higher than the corresponding rates in the IR group (4.24‰, 9.96‰, 4.04‰, and 1.98‰; log-rank test, all P < 0.05; Figure 2A-D).
Figure 2.
Kaplan-Meier survival curves for the cumulative incidence for ADs and NADs stratified by immune recovery status in the low-threshold cohort. Immune recovery status was categorized as INR ang IR. Panels A – D depict the probability of ADs: (A) PCP, (B) TB, (C) TSM, and (D) ADC. Panels E – H show the cumulative incidence of NADs: (E) ESLD, (F)CVD, (G) CKD, and (H) NADCs. The INR group demonstrated a higher cumulative incidence across most conditions compared with the IR group, except for TB. Abbreviations: ADs, AIDS-defining diseases; NADs, non-AIDS-defining diseases; INR, immune non-responders; IR, immune responders; PCP, pneumocystis pneumonia; TB, tuberculosis; TSM, talaromycosis marneffei; ADC, AIDS-defining cancers; ESLD, end-stage liver disease; CVD, cardiovascular disease; CKD, chronic kidney disease; NADC, non-AIDS-defining cancers.
In the high-threshold cohort, the median follow-up duration for the four ADs was 42.20–42.47 months (range). During these follow-up periods, 35 participants developed PCP, 46 developed TB, 32 developed TSM, and 12 developed ADC. The 5-year cumulative incidence rates of PCP, TB, TSM, and ADC in the INR group were 14.99‰, 13.84‰, 16.12‰, and 5.56‰, respectively, all higher than those observed in the IR (3.33‰, 7.82‰, 2.84‰, and 0.5‰). Except for TB, the differences in cumulative incidence rates for the other three diseases were statistically significant (log-rank test, all P < 0.05; Figure 2E-H).
Multivariate analysis of the association between INR and ADs
To investigate the impact of INR on the incidence of four ADs, both crude and adjusted COX models were constructed. In both cohorts, the crude model (Model 1) and adjusted models (Model 2 and 3) consistently showed a higher risk of PCP, TSM, and ADC in the INR group compared to the IR group. However, the increased risk of TB did not reach statistical significance. Specifically, in the low-threshold cohort, after adjusting for demographic characteristics, comorbidities, HIV-related factors, and laboratory parameters, the fully adjusted model (Model 3) showed that INR was associated with significantly increased risks of PCP (adjusted hazard ratio [aHR], 10.10; 95% CI: 4.94–20.70), TSM (aHR, 7.38; 95% CI: 3.51–15.50), and ADC (aHR, 3.67; 95% CI: 1.20–11.20) (Table 2). In the high-threshold cohort, INR was similarly associated with higher risks of PCP (aHR, 4.11; 95% CI: 1.97–8.58), TSM (aHR, 6.64; 95% CI: 2.87–15.40), and ADC (aHR, 4.39; 95% CI: 1.25–15.40) (Table 3). These results remained consistent in sensitivity analyses excluding individuals with less than six months of follow-up (Supplementary Table 11).
Table 2.
Associations between immune reconstitution status and incidence of ADs and NADs in the low-threshold cohort.
| State of immune reconstitution | Participants, n | No. of cases/Person-months | HR (95% CI) | ||
|---|---|---|---|---|---|
| Model1 | Model2 | Model3 | |||
| PCP | |||||
| IR | 5981 | 22/320387.90 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 861 | 17/36091.73 | 7.05 (3.74-13.30) | 7.34 (3.88-13.90) | 10.10 (4.94-20.70) |
| P Value | <0.001 | <0.001 | <0.001 | ||
| TB | |||||
| IR | 5770 | 53/218584.17 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 819 | 11/84763.27 | 1.81 (0.942-3.46) | 1.80 (0.936-3.46) | 1.97 (0.941-4.11) |
| P Value | 0.0750 | 0.078 | 0.072 | ||
| TSM | |||||
| IR | 5961 | 20/319488.83 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 847 | 15/35759.87 | 7.18 (3.67-14.1) | 7.12 (3.61-14.00) | 7.38 (3.51-15.50) |
| P Value | <0.001 | <0.001 | <0.001 | ||
| ADC | |||||
| IR | 5982 | 13/320606.00 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 857 | 5/36249.33 | 3.34 (1.19-9.40) | 2.99 (1.05-8.49) | 3.67 (1.20-11.20) |
| P Value | <0.05 | <0.05 | <0.05 | ||
| ESLD | |||||
| IR | 5982 | 9/320788.07 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 856 | 9/36109.93 | 8.97 (3.55-22.60) | 8.84 (3.48-22.40) | 15.00 (5.59-40.00) |
| P Value | <0.001 | <0.001 | <0.001 | ||
| CVD | |||||
| IR | 5982 | 52/319722.2 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 858 | 19/36000.6 | 3.36 (1.98-5.68) | 3.14 (1.85-5.33) | 3.83 (2.14-6.87) |
| P Value | <0.001 | <0.001 | <0.001 | ||
| CKD | |||||
| IR | 5982 | 210/318646.53 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 857 | 45/35968.53 | 2.08 (1.50-2.87) | 1.69 (1.22-2.34) | 1.78 (1.23-2.58) |
| P Value | <0.001 | <0.01 | <0.01 | ||
| NADC | |||||
| IR | 5982 | 29/320449.90 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 856 | 14/36072.97 | 4.49 (2.37-8.50) | 4.14 (2.18-7.90) | 4.75 (2.31-9.74) |
| P Value | <0.001 | <0.001 | <0.001 | ||
ADs: AIDS-defining diseases; NADs: non-AIDS-defining diseases; INR: immune non-responders; IR: immune responders; PCP: pneumocystis pneumonia; TB: tuberculosis; TSM: talaromycosis marneffei; ADC: AIDS-defining cancers; ESLD: end-stage liver disease; CVD: cardiovascular disease; CKD: chronic kidney disease; NADC: non-AIDS-defining cancers; HR: Hazard ratio; CI: Confidence interval
Model 1: An unadjusted model
Model 2: Adjusted for sex, age
Model 3: All diseases were adjusted for sex, age, marital status, body mass index, smoking, drinking, HIV transmission route, white blood cells, platelet, CD8 count, HIV RNA, HBV, HCV, WHOstage, opportunistic infections, the time between the diagnosis of HIV and the initiation of ART, ART regimen.
ESLD further adjusted for alanine aminotransferase, aspartate aminotransferase, total bilirubin and gamma-glutamyltransferase.
CVD further adjusted for alanine aminotransferase and aspartate aminotransferase.
CKD further adjusted for creatinine.
TSM further adjusted for aspartate aminotransferase.
Marital status, body mass index, smoking, drinking, WHOstage, opportunistic infections, ART treatment regimen were collected at the start of initial treatment.
Table 3.
Associations between immune reconstitution status and incidence of ADs and NADs in the high-threshold cohort.
| State of immune reconstitution | Participants, n | No. of cases/Person-months | HR (95% CI) | ||
|---|---|---|---|---|---|
| Model1 | Model2 | Model3 | |||
| PCP | |||||
| IR | 4849 | 15/226580.33 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 2030 | 20/88321.57 | 3.39 (1.74-6.63) | 3.40 (1.74-6.65) | 4.11 (1.97-8.58) |
| P Value | <0.001 | <0.001 | <0.001 | ||
| TB | |||||
| IR | 4682 | 29/218584.17 | 1.00(reference) | 1.00(reference) | 1.00(reference) |
| INR | 1947 | 17/84763.27 | 1.51 (0.828-2.74) | 1.49 (0.82-2.72) | 1.62 (0.843-3.13) |
| P Value | 0.179 | 0.190 | 0.147 | ||
| TSM | |||||
| IR | 4840 | 11/226244.80 | 1.00(reference) | 1.00(reference) | 1.00(reference) |
| INR | 2022 | 21/88094.07 | 5.00 (2.41-10.40) | 4.95 (2.39-10.30) | 6.64 (2.87-15.40) |
| P Value | <0.001 | <0.001 | <0.001 | ||
| ADC | |||||
| IR | 4848 | 4/226988.30 | 1.00(reference) | 1.00(reference) | 1.00(reference) |
| INR | 2028 | 8/88465.27 | 5.13 (1.55-17.10) | 5.11 (1.54-17.00) | 4.39 (1.25-15.40) |
| P Value | <0.01 | <0.01 | <0.05 | ||
| ESLD | |||||
| IR | 4848 | 4/227003.83 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 2027 | 8/88475.33 | 5.16 (1.55-17.20) | 5.10 (1.53-16.90) | 4.60 (1.32-16.10) |
| P Value | <0.01 | <0.01 | <0.05 | ||
| CVD | |||||
| IR | 4848 | 41/226037.0 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 2030 | 27/88100.3 | 1.68 (1.03-2.73) | 1.67 (1.03-2.72) | 1.95 (1.17-3.26) |
| P Value | <0.05 | <0.05 | <0.05 | ||
| CKD | |||||
| IR | 4848 | 148/225572.1 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 2027 | 96/87738.1 | 1.70 (1.31-2.20) | 1.70 (1.31-2.19) | 1.73 (1.31-2.28) |
| P Value | <0.001 | <0.001 | <0.001 | ||
| NADC | |||||
| IR | 4848 | 16/226593.8 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| INR | 2030 | 26/88373.8 | 4.15 (2.22-7.73) | 4.15 (2.23-7.74) | 4.17 (2.12-8.20) |
| P Value | <0.001 | <0.001 | <0.001 | ||
ADs: AIDS-defining diseases; NADs: non-AIDS-defining diseases; INR: immune non-responders; IR: immune responders; PCP: pneumocystis pneumonia; TB: tuberculosis; TSM: talaromycosis marneffei; ADC: AIDS-defining cancers; ESLD: end-stage liver disease; CVD: cardiovascular disease; CKD: chronic kidney disease; NADC: non-AIDS-defining cancers; HR: Hazard ratio; CI: Confidence interval
Model 1: an unadjusted model
Model 2: adjusted for sex, age
Model 3: All diseases were adjusted for sex, age, marital status, body mass index, smoking, drinking, HIV transmission route, white blood cells, platelet, CD8 count, HIV RNA, HBV, HCV, WHOStage, Opportunistic infections, the time between the diagnosis of HIV and the initiation of ART, ART regimen.
ESLD further adjusted for alanine aminotransferase, aspartate aminotransferase, total bilirubin and gamma-glutamyltransferase.
CVD further adjusted for alanine aminotransferase and aspartate aminotransferase.
CKD further adjusted for creatinine.
TSM further adjusted for aspartate aminotransferase.
Marital status, body mass index, smoking, drinking, WHOstage, opportunistic infections, ART treatment regimen were collected at the start of initial treatment.
Comparison of cumulative incidence of NADs between INR and IR PWH
In the low-threshold cohort, the median follow-up duration for the four NADs was 49.10–49.42 months (range). During these periods, 18 participants developed ESLD, 71 developed CVD, 255 developed CKD, and 43 developed NADC. The 5-year cumulative incidence rates of ESLD, CVD, CKD, and NADC in the INR group were 13.59‰, 35.15‰, 81.49‰, and 20.64‰, respectively, all significantly higher than those in the IR group (1.62‰, 8.55‰, 36.26‰, and 5.26‰; log-rank test, all P < 0.05; Figure 3A-D).
Figure 3.
Kaplan-Meier survival curves for the cumulative incidence for ADs and NADs stratified by immune recovery status in the high-threshold cohort. Immune recovery status was categorized as INR ang IR. Panels A – D depict the probability of ADs: (A) PCP, (B) TB, (C) TSM, and (D) ADC. Panels E – H show the cumulative incidence of NADs: (E) ESLD, (F)CVD, (G) CKD, and (H) NADCs. The INR group exhibited higher cumulative incidence across most conditions compared to the IR group, with no significant difference observed for TB. Abbreviations: ADs, AIDS-defining diseases; NADs, non-AIDS-defining diseases; INR, immune non-responders; IR, immune responders; PCP, pneumocystis pneumonia; TB, tuberculosis; TSM, talaromycosis marneffei; ADC, AIDS-defining cancers; ESLD, end-stage liver disease; CVD, cardiovascular disease; CKD, chronic kidney disease; NADC, non-AIDS-defining cancers.
In the high-threshold cohort, the median follow-up duration for the four NADs was 42.03–42.47 months (range). During these periods, 12 participants developed ESLD, 68 developed CVD, 244 developed CKD, and 42 developed NADC. The 5-year cumulative incidence rates in the INR group were 6.32‰ for ESLD, 16.29‰ for CVD, 73.77‰ for CKD, and 17.38‰ for NADC, all significantly higher than the corresponding rates in the IR group (0.94‰, 8.73‰, 41.41‰, and 4.22‰; log-rank test, all P < 0.05; Figure 3E-H).
Multivariate analysis of the association between INR and NADs
In both cohorts, the crude model (Model 1) and adjusted models (Model 2 and 3) consistently showed higher risk of NADs in the INR group compared to the IR group. Specifically, in the low-threshold cohort, the fully adjusted model (Model 3) indicated that INR was associated with significantly increased risks of ESLD (aHR, 15.00; 95% CI: 5.59–40.00), CVD (aHR, 3.83; 95% CI: 2.14–6.87), CKD (aHR, 1.78; 95% CI: 1.23–2.58), and NADC (aHR, 4.75; 95% CI: 2.31–9.74) (Table 2). Similarly, in the high-threshold cohort, INR was associated with elevated risks of ESLD (aHR, 4.60; 95% CI: 1.32–16.10), CVD (aHR, 1.95; 95% CI: 1.17–3.26), CKD (aHR, 1.73; 95% CI: 1.31–2.28), and NADC (aHR, 4.17; 95% CI: 2.12–8.20) (Table 3). These associations remained consistent in sensitivity analyses excluding individuals with less than six months of follow-up (Supplementary Table 12).
Sensitivity analyses
After excluding potential mediators identified through a DAG – such as opportunistic infections at baseline – the aHRs for both ADs and NADs remained materially unchanged (Supplementary Figure 2 and Supplementary Tables 13,14). In the IPTW analysis, baseline covariate balance between INR and IR groups was substantially improved, with all SMD reduced to below 0.1 (Supplementary Figures 3 and 4). The IPTW-aHRs were consistent with those from the primary multivariable models, supporting the robustness of the findings. E-values analyses indicated that it is unlikely that an unmeasured confounder could alter the observed associations (Supplementary Tables 15, 16). The associations remained robust when analysed using the Fine – Gray sub-distribution hazard models, which accounted for death and loss to follow-up as competing risks (Supplementary Tables 17 and 18). Corresponding CIF plots further illustrated the time-dependent divergence in risk between INR and IR groups (Supplementary Figures 5 and 6).
Discussion
This study provides a novel perspective by defining INR using two distinct CD4+ T-cell thresholds and examining its long-term impact. Our findings demonstrate that INR individuals remain at significantly higher risk for both opportunistic infections and chronic non-AIDS comorbidities compared to those who achieved robust immune recovery. These results underscore the long-term clinical consequences of INR and highlight INR as a critical determinant of adverse health outcomes in PWH.
There is currently no universally accepted definition of INR. Prior studies have employed varying CD4+ T-cell thresholds – ranging from 200 to 500 cells/µL – resulting in wide discrepancies in estimates INR prevalence, from 9% to 45% [17]. Mechanistically, INR has been linked to microbial translocation, residual viral reservoirs, lymphoid tissue fibrosis, and diminished thymic output. These immune impairments may be further exacerbated by host-related factors, including ageing and cytokine gene polymorphisms [18,19]. Collectively, these mechanisms contribute to incomplete immune reconstitution and sustained immunodeficiency, despite effective ART.
Aligned with this immunodeficient profile, our research observed that even after four or five years of ART, individuals with persistently low CD4+ T-cell counts (<350 or <500 cells/μL) remained at a significantly elevated risk of ADs, particularly PCP and TSM, compared to IRs. This pattern aligns with previous studies, reinforcing the long-term clinical consequences of suboptimal immune recovery [20,21]. Impaired Th1 function, defective IL-17 signalling, and compromised mucosal immunity may contribute to the persistent vulnerability of INR individuals to fungal infections [22-24]. These findings underscore that while ART promotes immune restoration and systemic immune balance, ADs remain a pressing concern in INR population, highlighting the importance of ongoing clinical vigilance for opportunistic infections in this group. Notably, our analysis revealed that a substantial proportion of ADs occurred in individuals whose CD4+ T-cell counts had risen above traditional immunodeficiency thresholds (≥350 or ≥500 cells/μL) at the time of diagnosis. This suggests that residual immune dysfunction – rather than absolute CD4+ T-cell depletion alone – may play a critical role in the pathogenesis of ADs. Such findings reinforce the need to evauate not only CD4+ T-cell quantity but also functional immune integrity in long-term ART care.
Although TB remains a major comorbidity in PWH, our study did not observe a statistically significant association between INR and TB incidence. This finding may reflect the complex immunopathological interplay between HIV and TB. In individuals with advanced HIV, the initiation of ART can precipitate immune reconstitution inflammatory syndrome – a dysregulated inflammatory response driven by rapid immune restoration – particularly in those with low baseline CD4+ T-cell counts and high mycobacterial burden [25,26]. Such paradoxical immune activation may lead to early TB events among IR. Conversely, INR individuals may experience insufficient CD4+ T-cell reconstitution to mount an inflammatory response strong enough to reactivate latent TB infection, thereby contributing to a lower observed incidence in this group.
Beyond opportunistic infections, our study provides compelling evidence that INR is also linked to a higher burden of NADs, particularly ESLD, CVD, and CKD. In this population, co-infection with HBV or HCV and ART-related hepatotoxicity synergistically to exacerbate persistent systemic inflammation, accelerating fibrogenesis and progression to cirrhosis through profibrotic cytokine pathways (e.g. TGF-β, IL-13) and immune-mediated hepatocyte injury [27-29]. Persistent immune activation and residual inflammation in INR individuals, manifested by elevated levels of IL-6, TNF-α, and soluble CD14, may also promote endothelial dysfunction, arterial stiffness, and atherosclerotic plaques instability, collectively increasing the risk of CVD events [30,31]. Notably, immune senescence may further exacerbate vascular injury by limiting endothelial repair capacity [32]. In the case of CKD, sustained immune dysregulation contributes to renal microvascular injury and progressive nephropathy, further aggravated by nephrotoxic ART regimens and metabolic comorbidities such as dyslipidemia and hypertension [33-35]. Accelerated immune senescence may also impair renal repair [36]. Together, these findings highlight that NADs represent a major component of the long-term disease burden in INR individuals and support a shift from CD4+ T cell-centric monitoring towards a multidimensional, risk-based surveillance strategy.
Notably, our findings further underscore the elevated risk of both ADCs and NADCs in INR individuals. While oncogenic viruses (e.g. HPV, HBV, EBV) are well-established drives of HIV-associated malignancies, accumulating evidence suggests that chronic immune dysregulation – independent of viral co-infections – also plays a critical role in oncogenesis in this population [37]. Severe CD4+ T-cell depletion compromises immune surveillance, increasing susceptibility to ADCs [38]. Simultaneously, persistent immune activation, systemic inflammation, and immune senescence may create a pro-tumorigenic microenvironment, even in the absence of traditional oncogenic viruses. Specifically, dysregulated cytokine signalling (e.g. IL-6, TNF-α), impaired antigen presentation, and reduced cytotoxic T-cell function may collectively impair tumour elimination and facilitate malignant transformation [39]. These processes are further exacerbated by behavioral risk factors, such as smoking and co-infections, which may synergize with immune dysfunction to amplify cancer risk, particularly in INR individuals [40-42]. Although our study lacked an HIV-negative control group, evidence from an Australian nationwide cohort and a large collaborative study from the NA-ACCORD consortium indicates that PWH – especially those with persistently low CD4+ T-cell counts – exhibit significantly higher NADC incidence than the general population [43, 44]. These results imply that incomplete immune recovery may have broader oncologic implications beyond viral oncogenesis, highlighting the need to elucidate the interplay between immune reconstitution, inflammation, and cancer biology. Such insights are critical to informing prevention, screening, and survivorship strategies tailored to INR individuals.
This study has several limitations. First, the exclusion of participants with incomplete data may have introduced potential selection bias. Second, although we adjusted for a wide range of covariates, residual confounding cannot ruled out, particularly from unmeasured factors such as dietary habits or genetic variations. Third, the absence of an HIV-negative comparator group limited our ability to contextualize the observed risks of ADs and NADs relative to the general population. Fourth, to better assess long-term morbidity, we excluded individuals who developed outcomes within the first four or five years of ART. While this approach helped isolate late-onset events, it may have introduced selection bias by excluding those with early disease progression or poor prognosis, potentially limiting the generalizability of our findings to the broader PWH population. Future studies should further evaluate the impact of early-event exclusion on risk estimates and cohort representativeness. Finally, as this was a single-centre study conducted in China, the results may not be fully generalizable to populations with different genetic backgrounds, healthcare systems, or environmental exposures.
In conclusion, this study provides robust evidence that INR substantially increases the long-term risk of both ADs and NADs in PWH. By employing a two-threshold definition of INR, we offer a novel perspective on the heterogeneity of immune recovery and its clinical consequences. Given the persistent vulnerability of INR individuals, reliance on CD4+ T-cell monitoring alone is insufficient. A multidimensional risk assessment framework is warranted – not only to incorporate inflammatory markers (e.g. IL-6, CD163, hsCRP) and immune activation indicators, but also to guide targeted clinical interventions such as cancer screening (e.g. for liver and anal cancers), vaccination programs (e.g. HPV and HBV), and proactive management of comorbidities (e.g. cardiovascular, metabolic, and neurocognitive disorders). Future research should prioritize multi-centre validation, refinement of long-term prognostic tools, and the implementation of precision-based strategies to reduce disease burden and improve outcomes in this high-risk population.
Author contributions
H. Z. L. and J. Y. L. conceived and designed the study. L. Q. S. and F. Z. contributed to data collection and curation. X. R. L. conducted formal analysis and wrote the first draft of the manuscript. Y. S. L., D. Z., Y. X. J., C. Y. L., H. W., and J. Y. H. developed the methodology. Y. H., H. Z. L. and J. Y. L. provided supervision. All authors critically reviewed and revised the final manuscript.
Supplementary Material
Acknowledgments
We are deeply grateful to all the participants enrolled and the clinical staff involved at the Shenzhen Third People's Hospital for their crucial contributions to this study.
Funding Statement
This work was supported by the project of the Guangdong Basic and Applied Basic Research Foundation [No. 2024A1515012118]，the Guangdong Provincial Medical Science and Technology Research Fund Project [No.A2025250], the Science and Technology Innovation Committee of Shenzhen Municipality [No. JCYJ20220531102202005], Shenzhen Clinical Reseach Center for Emerging Infectious Diseases Research Center for Emerging Infectious Diseases [No. LCYSSQ20220823091203007], Shenzhen High-level Hospital Construction Fund [No. G2022153，XKJS-GRMYK-001, XKJS-GRMYK-002], and Sanming Project of Medicine in Shenzhen [SZSM202311033].
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
All de-identified participant data analysed and presented in this study are available from the corresponding author following publication, on reasonable request.
Supplemental Material
Supplemental data for this article can be accessed online at https://doi.org/10.1080/22221751.2025.2539198.
References
- 1.Antiretroviral Therapy Cohort Collaboration . Survival of HIV-positive patients starting antiretroviral therapy between 1996 and 2013: a collaborative analysis of cohort studies. Lancet HIV. 2017;4(8):e349–e356. doi: 10.1016/S2352-3018(17)30066-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Yang X, Su B, Zhang X, et al. Incomplete immune reconstitution in HIV/AIDS patients on antiretroviral therapy: challenges of immunological non-responders. J Leukoc Biol. 2020;107(4):597–612. doi: 10.1002/JLB.4MR1019-189R [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Li H, Tang Y, Wang Y, et al. Single-cell sequencing resolves the landscape of immune cells and regulatory mechanisms in HIV-infected immune non-responders. Cell Death Dis. 2022;13(10):849), doi: 10.1038/s41419-022-05225-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bell LCK, Noursadeghi M.. Pathogenesis of HIV-1 and mycobacterium tuberculosis co-infection. Nat Rev Microbiol. 2018;16(2):80–90. doi: 10.1038/nrmicro.2017.128 [DOI] [PubMed] [Google Scholar]
- 5.Lodi S, Guiguet M, Costagliola D, et al. Kaposi sarcoma incidence and survival among HIV-infected homosexual men after HIV seroconversion. J Natl Cancer Inst. 2010;102(11):784–792. doi: 10.1093/jnci/djq134 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Phair J, Muñoz A, Detels R, et al. The risk of pneumocystis carinii pneumonia among men infected with human immunodeficiency virus type 1. multicenter AIDS cohort study group. N Engl J Med. 1990;322(3):161–165. doi: 10.1056/NEJM199001183220304 [DOI] [PubMed] [Google Scholar]
- 7.Phillips AN, Neaton J, Lundgren JD.. The role of HIV in serious diseases other than AIDS. AIDS. 2008;22(18):2409–2418. doi: 10.1097/QAD.0b013e3283174636 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ad M, Abrams D, Pradier C, et al. HIV-induced immunodeficiency and mortality from AIDS-defining and non-AIDS-defining malignancies. AIDS. 2008;22(16):2143–2153. doi: 10.1097/QAD.0b013e3283112b77 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Maartens G, Celum C, Lewin SR.. HIV infection: epidemiology, pathogenesis, treatment, and prevention. Lancet. 2014;384(9939):258–271. doi: 10.1016/S0140-6736(14)60164-1 [DOI] [PubMed] [Google Scholar]
- 10.Marcus JL, Leyden WA, Alexeeff SE, et al. Comparison of overall and comorbidity-free life expectancy between insured adults with and without HIV infection, 2000-2016. JAMA Netw Open. 2020;3(6):e207954. doi: 10.1001/jamanetworkopen.2020.7954 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Martínez-Sanz J, Díaz-Álvarez J, Rosas M, et al. Expanding HIV clinical monitoring: the role of CD4, CD8, and CD4/CD8 ratio in predicting non-AIDS events. EBioMedicine. 2023;95:104773. doi: 10.1016/j.ebiom.2023.104773 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Acquired Immunodeficiency Syndrome Professional Group Society of Infectious Diseases, Chinese Medical Association; Chinese Center for Disease Control and Prevention. Chinese guidelines for the diagnosis and treatment of human immunodeficiency virus infection/acquired immunodeficiency syndrome (2024 edition). Chin Med J (Engl). 2024;137(22):2654-2680. doi: 10.1097/CM9.0000000000003383 [DOI] [PMC free article] [PubMed]
- 13.Society of Infectious Diseases, Chinese Medical Association. Expert consensus on diagnosis and treatment of end-stage liver disease complicated with infections (2021 version) . [J]. J Clin Hepatol. 2022;38:304–310. doi: 10.3969/j.issn.1001-5256.2022.02.010 [DOI] [Google Scholar]
- 14.Hsue PY, Waters DD.. HIV infection and coronary heart disease: mechanisms and management. Nat Rev Cardiol. 2019;16(12):745–759. doi: 10.1038/s41569-019-0219-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease . Kidney Int. 2024;105(4S):S117 – S314. doi: 10.1016/j.kint.2023.10.018 [DOI] [PubMed] [Google Scholar]
- 16.Centers for Disease Control and Prevention (CDC) . Revised surveillance case definition for HIV infection – United States, 2014. MMWR Recomm Rep. 2014;63(RR-03):1–10. [PubMed] [Google Scholar]
- 17.Acquired Immunodeficiency Syndrome and Hepatitis C Professional Group, Society of Infectious Diseases, Chinese Medical Association. Consensus on diagnosis and management of immunological non-responder in acquired immunodeficiency syndrome (version 2023) . Chin J Infect Dis. 2024;42:3–13. doi: 10.3760/cma.j.cn311365-20230927-00098 [DOI] [Google Scholar]
- 18.Wan LY, Huang HH, Zhen C, et al. Distinct inflammation-related proteins associated with T cell immune recovery during chronic HIV-1 infection. Emerg Microbes Infect. 2023;12(1):2150566. doi: 10.1080/22221751.2022.2150566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lu D, Zhang JB, Wang YX, et al. Association between CD4+ T cell counts and gut microbiota and serum cytokines levels in HIV-infected immunological non-responders. BMC Infect Dis. 2021;21(1):742), doi: 10.1186/s12879-021-06491-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Narayanasamy S, Dat VQ, Thanh NT, et al. A global call for talaromycosis to be recognised as a neglected tropical disease. Lancet Glob Health. 2021;9(11):e1618 – e1622. doi: 10.1016/S2214-109X(21)00350-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Atkinson A, Miro JM, Mocroft A, et al. No need for secondary pneumocystis jirovecii pneumonia prophylaxis in adult people living with HIV from Europe on ART with suppressed viraemia and a CD4 cell count greater than 100 cells/µL. J Int AIDS Soc. 2021;24(6):e25726), doi: 10.1002/jia2.25726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zhang NN, Huang X, Feng HY, et al. Circulating and pulmonary T-cell populations driving the immune response in non-HIV immunocompromised patients with pneumocystis jirovecii pneumonia. Int J Med Sci. 2019;16(9):1221–1230. doi: 10.7150/ijms.34512 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Qiu Y, Li Z-T, Zeng W, et al. Th1 cell immune response in talaromyces marneffei infection with anti-interferon-γ autoantibody syndrome. Microbiol Spectr. 2024;12(5):e0364623. doi: 10.1128/spectrum.03646-23 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Dong RJ, Zhang YG, Zhu L, et al. Innate immunity acts as the major regulator in talaromyces marneffei coinfected AIDS patients: cytokine profile surveillance during initial 6-month antifungal therapy. Open Forum Infect Dis. 2019;6(6):ofz205. doi: 10.1093/ofid/ofz205 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Narendran G, Jyotheeswaran K, Senguttuvan T, et al. Characteristics of paradoxical tuberculosis-associated immune reconstitution inflammatory syndrome and its influence on tuberculosis treatment outcomes in persons living with HIV. Int J Infect Dis. 2020;98:261–267. doi: 10.1016/j.ijid.2020.06.097 [DOI] [PubMed] [Google Scholar]
- 26.Tibúrcio R, Barreto-Duarte B, Naredren G, et al. Dynamics of T-lymphocyte activation related to paradoxical tuberculosis-associated immune reconstitution inflammatory syndrome in persons with advanced HIV. Front Immunol. 2021;12:757843. doi: 10.3389/fimmu.2021.757843 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mocroft A, Geressu A, Beguelin C, et al. The role of HIV/hepatitis B virus/hepatitis C virus RNA+ triple infection in end-stage liver disease and all-cause mortality in Europe. AIDS. 2023;37(1):91–103. doi: 10.1097/QAD.0000000000003406 [DOI] [PubMed] [Google Scholar]
- 28.Xu M, Warner C, Duan X, et al. HIV coinfection exacerbates HBV-induced liver fibrogenesis through a HIF-1α – and TGF-β1-dependent pathway. J Hepatol. 2024;80(6):868–881. doi: 10.1016/j.jhep.2024.01.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ryom L, Lundgren JD, De Wit S, et al . Use of antiretroviral therapy and risk of end-stage liver disease and hepatocellular carcinoma in HIV-positive persons. AIDS. 2016;30(11):1731–1743. doi: 10.1097/QAD.0000000000001018 [DOI] [PubMed] [Google Scholar]
- 30.So-Armah K, Benjamin LA, Bloomfield GS, et al. HIV and cardiovascular disease. Lancet HIV. 2020;7(4):e279 – e293. doi: 10.1016/S2352-3018(20)30036-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Li X, Cao W, Liu Z, et al. Effect of long-term combination anti-retroviral therapy on cardiovascular disease risks in human immunodeficiency virus/acquired immunodeficiency syndrome patients. Chin J Infect Dis. 2022;40(8):496–504. doi: 10.3760/cma.j.cn311365-20211118-00404 [DOI] [Google Scholar]
- 32.Liu Z, Liang Q, Ren Y, et al. Immunosenescence: molecular mechanisms and diseases. Signal Transduct Target Ther. 2023;8(1):200. doi: 10.1038/s41392-023-01451-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Jourde-Chiche N, Fakhouri F, Dou L, et al. Endothelium structure and function in kidney health and disease. Nat Rev Nephrol. 2019;15(2):87–108. doi: 10.1038/s41581-018-0098-z [DOI] [PubMed] [Google Scholar]
- 34.Flandre P, Pugliese P, Cuzin L, et al. Risk factors of chronic kidney disease in HIV-infected patients. Clin J Am Soc Nephrol. 2011;6(7):1700–1707. doi: 10.2215/CJN.09191010 [DOI] [PubMed] [Google Scholar]
- 35.Gao H, Zhang J, Yang X, et al. The incidence and dynamic risk factors of chronic kidney disease among people with HIV. AIDS. 2023;37(12):1783–1790. doi: 10.1097/QAD.0000000000003662 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Sato Y, Yanagita M.. Immunology of the ageing kidney. Nat Rev Nephrol. 2019;15(10):625–640. doi: 10.1038/s41581-019-0185-9 [DOI] [PubMed] [Google Scholar]
- 37.Zhang W, Ruan L.. Recent advances in poor HIV immune reconstitution: what will the future look like? Front Microbiol. 2023;14:1236460. doi: 10.3389/fmicb.2023.1236460 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Chammartin F, Mocroft A, Egle A, et al. Measures of longitudinal immune dysfunction and risk of AIDS and non-AIDS defining malignancies in antiretroviral-treated people with human immunodeficiency virus. Clin Infect Dis. 2024;78(4):995–1004. doi: 10.1093/cid/ciad671 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Vos WAJW, Navas A, Meeder EMG, et al. HIV immunological non-responders are characterized by extensive immunosenescence and impaired lymphocyte cytokine production capacity. Front Immunol. 2024;15:1350065. doi: 10.3389/fimmu.2024.1350065 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Kirk GD, Merlo CA; Lung HIV Study . HIV infection in the etiology of lung cancer: confounding, causality, and consequences. Proc Am Thorac Soc. 2011;8(3):326-332. doi: 10.1513/pats.201009-061WR [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.He Y, Cai W, Chen J, et al. Persistent chronic immune activation in HIV/HBV-coinfected patients after antiretroviral therapy. J Viral Hepat. 2021;28(10):1355–1361. doi: 10.1111/jvh.13559 [DOI] [PubMed] [Google Scholar]
- 42.Chaillon A, Nakazawa M, Rawlings SA, et al. Subclinical cytomegalovirus and epstein-barr virus shedding is associated with increasing HIV DNA molecular diversity in peripheral blood during suppressive antiretroviral therapy. J Virol. 2020;94(19):e00927–20. doi: 10.1128/JVI.00927-20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wong IKJ, Grulich AE, Poynten IM, et al. Time trends in cancer incidence in Australian people living with HIV between 1982 and 2012. HIV Med. 2022;23(2):134–145. doi: 10.1111/hiv.13179 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Sun J, Althoff KN, Jing Y, et al. Trends in hepatocellular carcinoma incidence and risk among persons with HIV in the US and Canada, 1996-2015. JAMA Netw Open. 2021;4(2):e2037512. doi: 10.1001/jamanetworkopen.2020.37512 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All de-identified participant data analysed and presented in this study are available from the corresponding author following publication, on reasonable request.