A novel prediction model to evaluate the probability of CD4+/CD8+ cell ratio restoration in HIV-infected individuals : AIDS

Secondary Logo

Journal Logo

CLINICAL SCIENCE

A novel prediction model to evaluate the probability of CD4+/CD8+ cell ratio restoration in HIV-infected individuals

Li, Beia; Zhang, Leidanb; Liu, Yinga; Xiao, Jingb; Li, Cuilinb; Fan, Linac; Duan, Yujiaoa; Xiao, Jianga; Hao, Yud; Han, Junyand; Kong, Yaxiand; Zhao, Hongxina

Author Information
doi: 10.1097/QAD.0000000000003167

Abstract

Introduction

Even though combination antiretroviral therapy (cART) has shifted HIV from a deadly disease to a chronic illness, the gap in life expectancy persists between infected and uninfected individuals [1,2]. This gap is mainly due to an elevated risk of serious non-AIDS events (SNAEs) [3–5], including cardiovascular events, malignancies, renal and hepatic impairment, bone diseases, and neurocognitive disorders [6]. Higher grades of D-dimer, soluble CD14, C-reactive protein, and IL-6 are related to SNAEs occurrence in the HIV-positive population [7–9], but these biomarkers are susceptible to interference [10], cumbersome measured and nonroutine items that the clinical application is restricted.

Nowadays, increasing evidence has shown that a lower or inverted CD4+/CD8+ ratio value can better reflect the risk of SNAEs [11–13]. More importantly, compared with the plasma HIV RNA and the CD4+ cell counts, the ratio can accurately reflect the disease progression, response to therapy, morbidity, mortality, and overall condition of immune activation and dysfunction [14,15]. Therefore, it has become a frontier marker to evaluate the extent of immune recovery of HIV/AIDS patients with enduring therapy [16].

Despite long-term treatment, only 30% of individuals with stable virological control achieved CD4+/CD8+ ratio restoration (≥1) [17]. Based on its unique position in the disease process, a deeper cognition of the CD4+/ CD8+ ratio and the manipulation factors should be a core for future studies. On the contrary, previous reports are limited to cross-sectional studies. Although several studies assessed factors related to ratio recovery, none of the longitudinal designs established a complete predictive model of ratio restoration situation or given detailed parameters.

Therefore, we aimed to create and validate a novel prediction model to evaluate the probability of CD4+/ CD8+ ratio restoration among HIV-positive individuals on long-term cART. According to the forecast consequences, finding those at high risk of failed immune recovery then supplies additional interventions to optimize treatment and surveillance.

Methods

Study design and participants

For setting up this multivariable prediction model, 1980 HIV-infected patients were selected for the retrospective analysis, who initiated cART from 1 January 2013, to 30 December 2016, at Beijing Ditan Hospital, one of the largest AIDS designated hospitals in China, and obtained persistent virological suppression within 4 years of treatment.

CD4+/CD8+ ratio recovery was defined as a value at least 1, based on that a ratio less than 1 was related to T-cell activation, senescence, and apoptosis [16,18], and compared with previously published studies [14,15,19].

Participants included were if their ratio was less than 1 while cART was introduced, achieved virological suppression within 24 weeks after treatment, follow-up for at least 4 years (the visit time was calculated from the introduction of cART), and experienced persistent HIV viral load suppression (viral load < 200 copies/ml). Exclusion criteria were pregnant women n = 3, overseas patients n = 5, active opportunistic infection n = 19, incomplete baseline data n = 67, and those who died or transferred back to the local area during follow-up (died = 3, transferred out = 8). The syphilis is confirmed by Treponema pallidum particle agglutination and titers of the toluidine red unheated serum test.

The current study is under the Declaration of Helsinki and was approved by the ethical committee of Beijing Ditan Hospital, Capital Medical University. All clinical and laboratory data were used anonymously, and written informed consent was not required due to retrospective study using de-identified clinical information.

Demographic and clinical data

The following demographic and clinical variables were collected: age at AIDS diagnosis, sex, HIV transmission route, WHO-based clinical-stage, BMI, syphilis, hepatitis B virus, hepatitis C virus serostatus, HIV viral load, CD4+/CD8+ ratio, CD4+ and CD8+ T-cell counts at pre-cART, initial treatment regimen, and calendar year at baseline. The laboratory parameters covered were: hemoglobin (HGB), albumin (ALB), blood glucose (GLU), serum creatinine, blood urea nitrogen, neutrophil/lymphocyte, monocyte/neutrophil, monocyte/lymphocyte, and platelet/lymphocyte.

Statistical analysis

The patients were randomly assigned into derivation and validation cohort in 7: 3 ratio by Statistical Package for the Social Sciences (SPSS) software (version 26.0; SPSS Inc. Chicago, Illinois, USA). Predictors of ratio restoration that were discussed repeatedly in systematic reviews or studies and easily ascertained in different settings with various clinical experiences and part of the routine examinations of people with HIV (PWH) were explored. Meanwhile, a univariable analysis with Cox proportional hazards regression in the derivation cohort was conducted to discover other potential novel factors not previously reported.

Demographic and laboratory factors at baseline of the above two cohorts were compared using Student's t test. Continuous variables were expressed as the median and interquartile range (P25–P75) and analyzed using the nonparametric Mann–Whitney U test. Categorical variables were presented as counts and percentages and assessed via the chi-square test or Fisher's exact probability test. In addition, referring to clinical practice and previous reports, continuous variables HGB, ALB, GLU, viral load [20], CD4+/CD8+ ratio [11], and CD4+/CD8+ counts [17] were converted into categorical variables.

Significant predictors from univariable Cox proportional regression analysis were included to establish a practical recovery model in the derivation cohort. All potential factors were performed in the multivariable analysis and confirmed the proportional hazard assumption (P < 0.05). At the same time, the nomogram models were created utilizing RMS in the R package (version 4.0.5; R studio, Boston, Massachusetts, USA) to make the outcomes more visualized.

The predictive accuracy of the above model was evaluated with discrimination and calibration analysis and conducted internal verification. Discrimination was assessed by the area under the receiver operating characteristic (AUROC) curve, ranging from 0.5 to 1.0. Calibration degree was determined by 45° diagonal line, and relevant deviation above or below meant underprediction or overprediction, respectively.

Results

Participant characteristics

The derivation cohort contained 1410 patients, and the validation cohort involved 570 PWH. The characteristics of participants are presented in Table 1. In general, the observation indicators showed no statistically significant difference between the two cohorts (P > 0.05).

Table 1 - Demographic and clinical characteristics of the study population at the start of antiretroviral therapy (n = 1980).
Variables Derivation cohort, n = 1410 Validation cohort, n = 570 P value
Age at HIV diagnosis (years) 29 (25, 35) 29 (25, 35) 0.499
Sex (male) 1355 (96.1) 549 (96.3) 0.820
Route of HIV transmission
 MSM 1222 (86.7) 488 (85.6) 0.535
 Heterosexual 101 (7.2) 49 (8.6)
 Other/Unknown 87 (6.1) 33 (5.8)
WHO clinical stage
 I–II 1234 (87.5) 506 (88.8) 0.439
 III–IV 176 (12.5) 64 (11.2)
BMI (kg/m2) 21.5 (19.9, 23.7) 21.9 (20.2, 23.8) 0.100
Baseline HGB (g/l) 153.0 (143.0, 162.0) 154.0 (143.0, 162.0) 0.368
Baseline ALB (g/l) 48.4 (45.4, 50.5) 48.2 (45.7, 50.8) 0.836
Baseline GLU 5.83 (5.37, 6.12) 5.84 (5.37, 6.11) 0.967
Baseline Scr 71.6 (65.0, 79.0) 71.0 (65.0, 78.0) 0.235
Baseline BUN 4.45 (3.77, 5.24) 4.40 (3.68, 5.18) 0.372
Syphilis
 Negative 917 (65.0) 383 (67.2) 0.360
 Positive 493 (35.0) 187 (32.8)
HBsAg
 Negative 1298 (92.1) 533 (93.5) 0.267
 Positive 112 (7.9) 37 (6.5)
Anti-HCV
 Negative 1371 (97.2) 551 (96.7) 0.498
 Positive 39 (2.8) 19 (3.3)
Baseline HIV-1 RNA (copies/ml)
 ≤100000 966 (68.5) 396 (69.5) 0.675
 >100000 444 (31.5) 174 (30.5)
CD4+/CD8+ ratio at baseline
 <0.4 1022 (72.5) 414 (72.6) 0.175
 0.4–0.69 336 (23.8) 144 (25.3)
 ≥0.7 52 (3.7) 12 (2.1)
Baseline CD4+ cell counts (cells/μl)
 <200 353 (25.1) 146 (25.6) 0.848
 200–349 512 (36.3) 198 (34.7)
 350–499 364 (25.8) 156 (27.4)
 ≥500 181 (12.8) 70 (12.3)
Baseline CD8+ cell counts (cells/μl)
 ≤ 1000 711 (50.4) 289 (50.7) 0.911
 > 1000 699 (49.6) 281 (49.3)
Initial cART regimen
 ABC/TDF + 3TC + EFV/NVP 1353 (96.0) 549 (96.3) 0.810
 ZDV + 3TC + EFV/NVP 49 (3.5) 17 (3.0)
 TDF + 3TC + LPV/r 8 (0.5) 4 (0.7)
Calendar year at cART introduction
 2013–2014 508 (36.0) 213 (37.4) 0.575
 2015–2016 902 (64.0) 357 (62.6)
N/L 1.70 (1.20, 2.53) 1.72 (1.23, 2.56) 0.487
M/N 0.12 (0.09, 0.16) 0.12 (0.09, 0.16) 0.810
M/L 0.19 (0.15, 0.26) 0.20 (0.16, 0.27) 0.100
P/L 107.40 (81.60, 143.53) 110.60 (84.88, 145.43) 0.324
P value between normalization and nonnormalization groups in the derivation cohort and validation cohort. Data are presented as n (%), or median (interquartile range). 3TC, lamivudine; ABC, abacavir; ALB, albumin; Anti-HCV, antibody to hepatitis C virus; BUN, blood urea nitrogen; cART, combination antiretroviral therapy; EFV, efavirenz; GLU, blood glucose; HBsAg, hepatitis B surface antigen; HGB, haemoglobin; LPV/r, lopinavir/ritonavir; M/L, monocyte-lymphocyte ratio; M/N, monocyte-neutrophil ratio; N/L, neutrophil-lymphocyte ratio; NVP, nevirapine; P/L, platelet-lymphocyte ratio; Scr, Serum creatinine; TDF, tenofovir disoproxil fumerate; ZDV, zidovudine.

Development of predictive model of the ratio restoration for people with HIV

After 4 years oftreatment, a total of 455 patients (22.98%) restored their ratio. Cox proportional hazards regression analysis was conducted to assess the effect of clinical factors on the ratio recovery of PWH.

Univariable analysis of the derivation cohort identified some significant previously reported factors: baseline viral load/CD4+ cell counts/CD8+ cell counts and pre-cART CD4+/CD8 + ratio, and several novel latent candidates were also revealed. These variables were included in a multivariate analysis, and finally, five independent predictors (age, ALB, syphilis, baseline CD4+ cell counts, and CD8+ cell counts) of ratio restoration were determined (Table 2).

Table 2 - Cox proportional hazard regression analysis to identify factors associated with CD4+/CD8+ ratio ≥1.0 from the derivation cohort.
Variables Univariate analysis Multivariate analysis
HR (95% CI) P value HR (95% CI) P value
Age at HIV diagnosis (years) 1.01 (0.99, 1.02) 0.084 1.03 (1.02, 1.04) <0.001
Sex
 Male 1
 Female 1.11 (0.64, 1.93) 0.713
Route of HIV transmission
 MSM 1
 Others 1.17 (0.86, 1.59) 0.328
WHO clinical stage
 i–ii 1
 III–IV 0.37 (0.23, 0.60) <0.001
BMI (kg/m2) 1.04 (1.00, 1.07) 0.026
Baseline HGB (g/l)
 ≤120 1
 >120 4.76 (2.12, 10.68) <0.001
Baseline ALB (g/l)
 ≤40 1
 >40 6.10 (2.72, 13.69) <0.001 2.64 (1.13, 6.15) 0.025
Baseline GLU
 <7.0 1
 ≥7.0 1.442 (0.990, 2.101) 0.057
Baseline Scr 1.003 (0.993, 1.014) 0.512
Baseline BUN 1.099 (1.008, 1.199) 0.033
Syphilis
 Negative 1
 Positive 0.79 (0.62, 1.00) 0.049 0.78 (0.61, 0.99) 0.046
HBsAg
 Negative 1
 Positive 1.13 (0.76, 1.66) 0.552
Anti-HCV
 Negative 1
 Positive 1.01 (0.52, 1.96) 0.980
Baseline HIV-1 RNA (copies/ml)
 ≤100000 1
 >100000 0.52 (0.40, 0.69) <0.001
CD4+/CD8+ ratio at baselinea
 <0.4 1
 0.4–0.69 6.01 (4.72, 7.66) <0.001
 ≥0.7 13.74 (9.51, 19.86) <0.001
Baseline CD4+ cell counts (cells/μl)
 <200 1
 200–349 8.38 (4.24, 16.57) <0.001 9.46 (4.70, 19.07) <0.001
 350–499 14.40 (7.31, 28.39) <0.001 20.79 (10.31, 41.92) <0.001
 ≥500 27.71 (13.96, 55.01) <0.001 53.28 (26.09, 108.80) <0.001
Baseline CD8+ cell counts (cells/μl)
 ≤1000 1
 >1000 0.64 (0.51, 0.80) <0.001 0.29 (0.23, 0.36) <0.001
Initial cART regimen
 ABC/TDF + 3TC + EFV/NVP 1
 ZDV + 3TC + EFV/NVP 0.79 (0.41, 1.53) 0.483
 TDF + 3TC + LPV/r 0.54 (0.08, 3.81) 0.533
Calendar year at cART introduction
 2013–2014 1
 2015–2016 1.18 (0.93, 1.49) 0.179
 N/L 0.91 (0.84, 0.99) 0.023
 M/N 0.02 (0.01, 0.18) <0.001
 M/L 0.11 (0.04, 0.30) <0.001
 P/L 0.997 (0.995, 0.999) <0.001
Data are presented as n (%), or median (interquartile range). Multivariate analysis is adjusted for all variables with P < 0.10 in Univariate analysis.3TC, lamivudine; ABC, abacavir; ALB, albumin; Anti-HCV, antibody to hepatitis C virus; BUN, blood urea nitrogen; cART, combination antiretroviral therapy; CI, confidence interval; EFV, efavirenz; GLU, blood glucose; HBsAg, hepatitis B surface antigen; HGB, hemoglobin; HR, hazard ratio; LPV/r, lopinavir/ritonavir; M/L, monocyte-lymphocyte ratio; M/N, monocyte-neutrophil ratio; N/L, neutrophil-lymphocyte ratio; NVP, nevirapine; P/L, platelet-lymphocyte ratio; Scr, Serum creatinine; TDF, tenofovir disoproxil fumerate; ZDV, zidovudine.
aThere was collinearity between baseline CD4+ cell counts and CD4+/CD8+ ratio, and so CD4+/CD8+ ratio was not included in the multivariable model.

A predictive equation was established via Cox regression analysis: ratio recovery index = 0.027 × Age + 0.969 × ALB–0.249 × Syphilis + (–/2.247/3.035/3.976) × CD4+ cell counts −1.252 × CD8+ cell counts (0 for patients without Syphilis, ALB level ≤ 40 g/l, CD4+ cell counts < 200 cells/μl, and CD8 + cell counts ≤1000 cells/μl; 1 for patients with Syphilis, ALB level > 40 g/l, CD4+ cell counts of 200–349 cells/μl, and CD8+ cell counts > 1000 cells/μl; 2 for patients with baseline CD4+ cell counts of 350–499 cells/μl; 3 for patients with CD4+ cell counts ≥500 cells/μl).

Discrimination of the ratio restoration model

According to the calculation results of the above equation, the study population was divided into the advantage group (>3.561) and the disadvantage group (≤3.561) with cut off values of 3.561. Compared with the disadvantage group, the advantage group had a significantly higher probability of ratio restoration both in the deriving and validation cohorts (P < 0.001) (Supplement Fig. 1A and B, https://links.lww.com/QAD/C437).

The AUROC was used to estimate the model's prediction capability, and the outcomes exhibited that AUROC in the derivation and validation cohort was 0.782 [95% confidence interval (CI): 0.760–0.804] and 0.743 (95% CI: 0.705–0.778), respectively. The distinction degree of the novel model notably acted better than the single parameters in two cohorts (derivation cohort, age: 0.520, ALB: 0.550, Syphilis: 0.530, CD4+ cell counts: 0.733, and CD8+ cell counts 0.562; validation cohort, age: 0.565, ALB: 0.542, Syphilis: 0.522, CD4+ cell counts: 0.691, and CD8+ cell counts 0.570) (Fig. 1a and b, Supplement Table 1, https://links.lww.com/QAD/C437).

F1
Fig. 1:
Receiver operating characteristic curves of the different variables for predicting the likelihood of CD4+/CD8+ ratio restoration in HIV/AIDS patients.

Calibration and visualization of the ratio restoration model

Calibration plots indicated well fitter between the actual observation and prediction data (Fig. 2a and b).

F2
Fig. 2:
Calibration plots and Nomogram for predicting CD4+/CD8+ ratio restoration in long-term treated HIV-infected individuals.

The prediction model was visualized further via a nomogram that combined five candidates determined by the Cox multivariable analyses. Besides the timepoint of the 4th year, our nomograms also offered feasible predictions at any time ranging from 1 to 4 years after successful viral suppression (Fig. 2c).

The decision curve analysis (DCA) was developed to compare three essential predictors’ clinical net benefit rate: the ratio restoration model, baseline CD4+ cell counts, and CD8+ levels, just as the final result displayed that the recovery model performed best (Fig. 3a and b).

F3
Fig. 3:
Decision curve analyses show the clinical benefit of the different indexes.

Discussion

The current observational cohort study established a practical tool to predict the CD4+/CD8+ ratio restoration (≥1) rate in HIV-infected individuals who achieved persistent viral load suppression with 4 years of antiretroviral therapy. Five independent indicators were involved: age at HIV confirmed, ALB condition, syphilis status, baseline CD4+ cell counts, and CD8+ levels. The prediction accuracy results demonstrated that the novel model had excellent discrimination and calibration, while the internal verification reflected its good applicability and extrapolation.

As expected, only a few patients experienced recovered CD4+/CD8+ ratio following a long period of stable cART (22.98% by 4 years), which may reflect persistent immune activation in chronic viral infection [11]. A low or inverted CD4+/CD8+ ratio is an immune risk phenotype linked closely with worsened immune function, immunosenescence, and chronic inflammation among HIV-1-infected individuals [13,21]. In general, the value of CD4+/CD8+ ratio increase is due to the improvement of CD4+ levels and normalization of CD8+ cell counts after viral suppression [17], while the most significant changes of ratio trajectories are driven by the latter [22]. Unluckily, CD8+ cells remain abnormal despite effective long-term treatment [23], and the details that influence the expansion of CD8+ quantities are unclear [24]. The imbalance modulation in early HIV infection and ongoing activation in later CD8+ cells may help understand why thorough reversion of the immunological dysfunction is not easy [25].

Our research and others demonstrated that a higher baseline CD4+/CD8+ ratio, CD4+ cell counts, and lower CD8+ levels were associated with an increased opportunity to obtain ratio recovery [26]. Further highlighted the significance of cART introduction at the early stage of HIV infection, while the CD4+ cell counts are high [17], which may lead to an increased number of patients who achieve normalization and ultimately benefit themselves [27].

In our result, the younger the age during AIDS diagnosis, the lower the probability of ratio restoration. This inconsistency of age effect was also reported in other studies [14,17,18]. One possible interpretation was related to the artifacts of replicative senescence in CD8+ cells. During aging, the renewal of naive CD8+ cells decreases with that the terminally differentiated CD8+ cells without proliferation capacity are enriched [28,29]. Does this make the inversion of the CD4+/ CD8+ ratio easier to rectify? Meanwhile, premature HIV infection could dramatically affect the functionality of the thymus, necessary for the achievement of complete immune recovery [30]. Furthermore, socioeconomic conditions may also contribute to this divergence [31].

Although baseline T-cell status and age have been associated with the response of ratio restoration, ALB levels were not reported as such a variable. At present, several studies imply that markers of nutritional condition, such as BMI, ALB, and HGB, may be used as risk indicators for the prognosis of HIV/AIDS patients [32], and that hypoproteinemia is usually one of the typical characteristics of cachexia [33].

The connection between HIV viremia and syphilis is multiplex and remains unclear. Syphilitic genital ulcers, a breach in the protective epithelial barrier, may create a portal for virus entry [34]. The inrush of plenty of macrophages and activated lymphocytes into ulcers then offers a cellular milieu rich in immune cells expressing receptors for the virus, leading to an increased odds for HIV acquisition [35]. In addition, syphilis infection may enhance the immune activation and cytokine secretion of host cells and thus accelerate HIV replication along with depressing CD4+ cell counts [36,37]. The features suggest that it is not surprising that the co-infected patients have a poor normalization rate. The latest data showed that after ART initiation, patients co-infected with HIV/syphilis had a smaller increase in CD4+ cell counts and CD4+/ CD8+ ratio than patients infected only with HIV [38].

To the best of our knowledge, this is the first step towards more personalized medicine because it could identify patients at high risk of failed immune restoration, and that more scientific and comprehensive medical management strategies are urgently needed. For high-hazard people, what are suitable methods to improve the CD4+/CD8+ ratio level, such as strengthening exercise and ETOH cessation? Should patients with syphilis receive extra visits to monitor T–cell subsets? Is there a benefit for ALB supplementation in daily life? These details need to be considered.

In addition, a nomogram can directly display the prediction odds of a single clinical factor and the whole model to advance clinical decision-making that is used widely in modern medical studies. The five variables identified by our analysis can be easily obtained from routine clinical settings, while the calibration plot was given to appraise how close the nomogram projected risk is to the actual risk.

Finally, the model was sufficiently powered to show favorable discrimination in the process of derivation and validation, respectively. It can be demonstrated by the AUROC, which exceeded 0.7 in two cohorts. In recent years, the DCA was proposed to evaluate further the clinical benefit of the prediction model by the clinical net benefit rate. By calculating DCA, our model based on the nomogram had a fine property compared with classic indicators (baseline CD4+ cell counts and CD8+ levels).

Despite the above strengths, we acknowledge several limitations. First, owing to the restriction of single-center studies, external verification could not be performed. Second, syphilis infection is prone to recurrence and reinfection [38]. Most patients are latent syphilis in clinical, and retrospective analysis cannot trace the epidemiological history. Therefore, it is difficult to determine the specific time of infection, which seriously hinders the analysis of the impact of prior syphilis, active syphilis, and individual host factors on the immune status of HIV/AIDS patients. In addition to the above, the lack of follow-up data (recommended at 1, 3, 6, 12, 24 months after syphilis treatment with benzathine penicillin G) also obstructs further exploration of why co-infected patients do not normalize effectively.

Furthermore, due to the constraints of retrospective research data, some parameters that may affect systemic inflammatory response and immune activation situation were not incorporated, such as cytomegalovirus co-infection [39], smoking mode [40], and alcohol abuse [41]. In our cohort, the median age of the participants was 29 years old ((interquartile range: 25–36) and was overwhelmed with male sex and MSM (populations with higher rates of cytomegalovirus and syphilis seropositivity). Therefore, the model's applicability among women and elderly HIV/AIDS patients needs to be further verified.

Our results also suggested that in clinical work, it was necessary to strengthen the early diagnosis of acute HIV infections and chronic infections, develop more sensitive and specific rapid detection reagents, implement joint detection of syphilis and HIV, promote multiple disciplinary collaboration, conduct regular screening, and case management of high-risk groups, timely preexposure prophylaxis and post-exposure prophylaxis, and then promote the prevention and treatment of HIV infection.

In the future, if the unique position of the CD4+/CD8+ ratio was confirmed in immunological interventions and HIV cure studies, the ratio restoration model might be a convenient tool to identify HIV/AIDS individuals most likely to benefit from modern antiretroviral drugs and immune therapy.

Acknowledgements

The authors acknowledge the work of HIV healthcare providers for their diagnosis, nursing, and treatment of HIV/AIDS patients in Ditan Hospital.

Author contributions: H.X.Z. conceived and designed the experiments; B.L. collected and analyzed the data and wrote the aticle; L.D.Z., Y.L., J.X., C.L.L., L.N.F., and YJ.D. provided patients data; J.X., Y.H., J.Y.H., and Y.X.K. analyzed the data; H.X.Z. approved the final version.

The current work was supported by the Beijing Municipal Administration of Hospitals’ Ascent Plan (no. DFL20191802) and Beijing Municipal Administration of Hospitals Clinical Medicine Development of Special Funding Support (no. ZYLX202126).

Conflicts of interest

All authors declared that there are no conflicts of interest.

References

1. McBride JA, Striker R. Imbalance in the game of T cells: what can the CD4/CD8 T-cell ratio tell us about HIV and health?. PLoS Pathog 2017; 13:e1006624.
2. Bourgi K, Wanjalla C, Koethe JR. Inflammation and metabolic complications in HIV. Curr HIV/AIDS Rep 2018; 15:371–381.
3. Smith CJ, Ryom L, Weber R, Morlat P, Pradier C, Reiss P, et al. Trends in underlying causes of death in people with HIV from 1999 to 2011 (D:A:D): a multicohort collaboration. Lancet 2014; 384:241–248.
4. Zicari S, Sessa L, Cotugno N, Ruggiero A, Morrocchi E, Concato C, et al. Immune activation, inflammation, and non-AIDS comorbidities in HIV-infected patients under long-term ART. Viruses 2019; 11:200.
5. Wada NI, Jacobson LP, Margolick JB, Breen EC, Macatangay B, Penugonda S, et al. The effect of HAART-induced HIV suppression on circulating markers of inflammation and immune activation. AIDS 2015; 29:463–471.
6. Perazzo H, Cardoso SW, Yanavich C, Nunes EP, Morata M, Gorni N, et al. Predictive factors associated with liver fibrosis and steatosis by transient elastography in patients with HIV mono-infection under long-term combined antiretroviral therapy. J Int AIDS Soc 2018; 21:e25201.
7. Borges AH, Silverberg MJ, Wentworth D, Grulich AE, Fätken-heuer G, Mitsuyasu R, et al. Predicting risk of cancer during HIV infection: the role of inflammatory and coagulation biomarkers. AIDS 2013; 27:1433–1441.
8. Nordell AD, McKenna M, Borges ÁH, Duprez D, Neuhaus J, Neaton JD, et al. Severity of cardiovascular disease outcomes among patients with HIV is related to markers of inflammation and coagulation. J Am Heart Assoc 2014; 3:e000844.
9. Tien PC, Choi AI, Zolopa AR, Benson C, Tracy R, Scherzer R, et al. Inflammation and mortality in HIV-infected adults: analysis of the FRAM study cohort. J Acquir Immune Defic Syndr 2010; 55:316–322.
10. Sullivan ZA, Wong EB, Ndung’u T, Kasprowicz VO, Bishai WR. Latent and active tuberculosis infection increase immune activation in individuals co-infected with HIV. EBioMedicine 2015; 2:334–340.
11. Castilho JL, Shepherd BE, Koethe J, Turner M, Bebawy S, Logan J, et al. CD4+/CD8+ ratio, age, and risk of serious noncommunicable diseases in HIV-infected adults on antiretroviral therapy. AIDS 2016; 30:899–908.
12. Serrano-Villar S, Pérez-Elias MJ, Dronda F, Casado JL, Moreno A, Royuela A, et al. Increased risk of serious non-AIDS-related events in HIV-infected subjects on antiretroviral therapy associated with a low CD4/CD8 ratio. PLoS One 2014; 9:e85798.
13. Mussini C, Lorenzini P, Cozzi-Lepri A, Lapadula G, Marchetti G, Nicastri E, et al. CD4/CD8 ratio normalisation and non-AIDS-related events in individuals with HIV who achieve viral load suppression with antiretroviral therapy: an observational cohort study. Lancet HIV 2015; 2:e98–e106.
14. Davy-Mendez T, Napravnik S, Zakharova O, Kuruc J, Gay C, Hicks CB, et al. Acute HIV infection and CD4/CD8 ratio normalization after antiretroviral therapy initiation. J Acquir Immune Defic Syndr 2018; 79:510–518.
15. Trickey A, May MT, Schommers P, Tate J, Ingle SM, Guest JL, et al. CD4:CD8 ratio and CD8 count as prognostic markers for mortality in human immunodeficiency virus-infected patients on antiretroviral therapy: the antiretroviral therapy cohort collaboration (ART-CC). Clin Infect Dis 2017; 65:959–966.
16. Lu W, Mehraj V, Vyboh K, Cao W, Li T, Routy J-P. CD4:CD8 ratio as a frontier marker for clinical outcome, immune dysfunction and viral reservoir size in virologically suppressed HIV-positive patients. J Int AIDS Soc 2015; 18:20052.
17. Caby F. HIV wcotCCRWGotFHDo,. CD4+/CD8+ ratio restoration in long-term treated HIV-1-infected individuals. AIDS 2017; 31:1685–1695.
18. Hove-Skovsgaard M, Zhao Y, Tingstedt JL, Hartling HJ, Thudium RF, Benfield T, et al. Impact of age and HIV status on immune activation, senescence and apoptosis. Front Immunol 2020; 11:583569.
19. Hughes RA, May MT, Tilling K, Taylor N, Wittkop L, Reiss P, et al. Long terms trends in CD4R cell counts, CD8+ cell counts, and the CD4+: CD8+ ratio. AIDS 2018; 32:1361–1367.
20. Smith CL, Stein GE. Viral load as a surrogate end point in HIV disease. Ann Pharmacother 2002; 36:280–287.
21. Serrano-Villar S, Sainz T, Lee SA, Hunt PW, Sinclair E, Shacklett BL, et al. HIV-infected individuals with low CD4/CD8 ratio despite effective antiretroviral therapy exhibit altered T cell subsets, heightened CD8+ T cell activation, and increased risk of non-AIDS morbidity and mortality. PLoS Pathog 2014; 10:e1004078.
22. Serrano-Villar S, Martínez-Sanz J, Ron R, Talavera-Rodríguez A, Fernández-Felix BM, Herrera S, et al. Effects of first-line anti-retroviral therapy on the CD4/CD8 ratio and CD8 cell counts in CoRIS: a prospective multicentre cohort study. Lancet HIV 2020; 7:e565–e573.
23. Helleberg M, Kronborg G, Ullum H, Ryder LP, Obel N, Gerstoft J. Course and clinical significance of CD8+ T-cell counts in a large cohort of HIV-infected individuals. J Infect Dis 2015; 211:1726–1734.
24. Nasi A, Chiodi F. Mechanisms regulating expansion of CD8+ T cells during HIV-1 infection. J Intern Med 2018; 283:257–267.
25. Cao W, Mehraj V, Trottier B, Baril J-G, Leblanc R, Lebouche B, et al. Early initiation rather than prolonged duration of antiretroviral therapy in HIV infection contributes to the normalization of CD8 T-cell counts. Clin Infect Dis 2016; 62:250–257.
26. Leung V, Gillis J, Raboud J, Cooper C, Hogg RS, Loutfy MR, et al. Predictors of CD4:CD8 ratio normalization and its effect on health outcomes in the era of combination antiretroviral therapy. PLoS One 2013; 8:e77665.
27. Thornhill J, Inshaw J, Oomeer S, Kaleebu P, Cooper D, Ramjee G, et al. Enhanced normalisation of CD4/CD8 ratio with early antiretroviral therapy in primary HIV infection. J Int AIDS Soc 2014; 17: (4 Suppl 3): 19480.
28. Dock JN, Effros RB. Role of CD8 T cell replicative senescence in human aging and in HIV-mediated immunosenescence. Aging Dis 2011; 2:382–397.
29. Desai S, Landay A. Early immune senescence in HIV disease. Curr HIV/AIDS Rep 2010; 7:4–10.
30. Bandera A, Ferrario G, Saresella M, Marventano I, Soria A, Zanini F, et al. CD4+ T cell depletion, immune activation and increased production of regulatory T cells in the thymus of HIV-infected individuals. PLoS One 2010; 5:e10788.
31. Petoumenos K, Choi JY, Hoy J, Kiertiburanakul S, Ng OT, Boyd M, et al. CD4:CD8 ratio comparison between cohorts of HIVpositive Asians and Caucasians upon commencement of antiretroviral therapy. Antivir Ther 2017; 22:659–668.
32. Sun Y, Luo J, Qian C, Luo L, Xu M, Min H, et al. The value of nutritional status in the prognostic analysis of patients with AIDS-related lymphoma. Infect Drug Resist 2021; 14:1105–1113.
33. Qin Y, Zhou Y, Lu Y, Chen H, Jiang Z, He K, et al. Multicentre derivation and validation of a prognostic scoring system for mortality assessment in HIV-infected patients with talaromycosis. Mycoses 2021; 64:203–211.
34. Stamm WE, Handsfield HH, Rompalo AM, Ashley RL, Roberts PL, Corey L. The association between genital ulcer disease and acquisition of HIV infection in homosexual men. JAMA 1988; 260:1429–1433.
35. Hook EW. Syphilis. Lancet 2017; 389:1550–1557.
36. Bentwich Z, Maartens G, Torten D, Lal AA, Lal RB. Concurrent infections and HIV pathogenesis. AIDS 2000; 14:2071–2081.
37. Palacios R, Jiménez-Oñate F, Aguilar M, Galindo MJ, Rivas P, Ocampo A, et al. Impact of syphilis infection on HIV viral load and CD4 cell counts in HIV-infected patients. J Acquir Immune Defic Syndr 2007; 44:356–359.
38. Fan L, Li C, Zhao H. Prevalence and risk factors of cytopenia in HIV-infected patients before and after the initiation of HAART. Biomed Res Int 2020; 2020:3132589.
39. Ramendra R, Isnard S, Lin J, Fombuena B, Ouyang J, Mehraj V, et al. Cytomegalovirus seropositivity is associated with increased microbial translocation in people living with human immunodeficiency virus and uninfected controls. Clin Infect Dis 2020; 71:1438–1446.
40. Valiathan R, Miguez MJ, Patel B, Arheart KL, Asthana D. Tobacco smoking increases immune activation and impairs T-cell function in HIV infected patients on antiretrovirals: a cross-sectional pilot study. PLoS One 2014; 9:e97698.
41. Monnig MA. Immune activation and neuroinflammation in alcohol use and HIV infection: evidence for shared mechanisms. Am J Drug Alcohol Abuse 2017; 43:7–23.
Keywords:

antiretroviral therapy; CD4+/CD8+ ratio; HIV; immune recovery; prediction model

Supplemental Digital Content

Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc.