A retrospective cohort study of Western blot band changes after treatment in 401 HIV-1 confirmed positive cases

Introduction

Western blot (WB) is widely considered the gold standard for confirming human immunodeficiency virus (HIV) antibodies and plays a critical role in Acquired Immunodeficiency Syndrome (AIDS) prevention and control (1-3). The characteristic WB bands primarily consist of ten specific proteins encoded by three genes: envelope protein antibodies (gp160, gp120, and gp41) encoded by the env gene; endonuclease and polymerase antibodies (p66, p51, and p31) encoded by the pol gene; and core protein antibodies (p17, p24, p39, and p55) encoded by the gag gene (4). The presence and combination of these bands directly influence the interpretation of WB antibody confirmation tests (5,6).

In previous follow-up studies, we observed a phenomenon in which serum antibody bands on WB appeared and subsequently disappeared in individuals with extremely early HIV infection who initiated antiretroviral therapy (ART) promptly, as well as in patients at the end stage of AIDS (7). This dynamic change can complicate the interpretation of WB results and increase the likelihood of indeterminate or even negative HIV antibody outcomes (1,7). Therefore, understanding the impact of ART on WB band changes and the identifying factors influencing band disappearance is particularly important for accurate HIV detection. Although several cohort studies have described HIV antibody dynamics in specific populations (e.g., children), most longitudinal WB investigations have focused on untreated individuals or early infection. Data on long-term WB band changes among adults receiving continuous ART, especially with irregular self-medication, remain limited. Additionally, Existing cross-sectional studies (8-10) are limited by single time-point assessment, narrow study populations (e.g., early infection, untreated individuals), and lack of clinical covariates (e.g., ART duration, age, viral load). They cannot capture dynamic WB band changes or establish associations between treatment-related factors and band evolution. In contrast, this longitudinal cohort study addresses these gaps by analyzing paired WB results over varying ART durations in a large adult cohort, with detailed demographic and clinical data.

In this study, we analyzed and compared initial WB patterns and second WB patterns after different treatment durations in 401 HIV-infected individuals diagnosed in Jiangsu Province. We explored the effects of treatment duration, demographic characteristics, and sexual transmission routes on WB band changes. We present this article in accordance with the STROBE reporting checklist (available at https://jphe.amegroups.com/article/view/10.21037/jphe-2026-0020/rc).

Methods

Study subjects

A retrospective observational study was performed among HIV-1-infected patients who initiated ART in Suzhou, Yangzhou, and Wuxi between January 1, 2010 and December 31, 2021. A total of 401 HIV-1-positive patients receiving standardized ART were enrolled based on predefined inclusion and exclusion criteria. To evaluate the effects of treatment duration on antibody dynamics, a stratified random sampling strategy was implemented based on predefined treatment durations (1, 3, 5, and 10 years), yielding an initial sample pool of 480 individuals. The inclusion criteria required that all participants were confirmed HIV-1 positive at baseline according to uniform national diagnostic standards and were receiving continuous, individualized ART. Subjects were excluded from the final analytical cohort if they met any of the following exclusion criteria: (I) missing or incomplete baseline WB original band records; or (II) missing follow-up clinical data or follow-up blood samples. Consequently, the final analytical cohort comprised 401 patients with complete paired baseline and follow-up data. The detailed screening, inclusion, and exclusion process is illustrated in Figure 1. Before December 31, 2021, whole blood samples of these 401 cases were collected using blood collection tubes containing ethylene diamine tetraacetic acid (EDTA) anticoagulant. Blood samples were processed in local laboratories (centrifuged at 3000 revolutions per minute for 15 minutes for separation) to obtain plasma samples, which were then uniformly sent to the provincial Center for Disease Control and Prevention laboratory.

Figure 1 Flowchart of the screening, inclusion, and exclusion process of participants. ART, antiretroviral therapy; HIV, human immunodeficiency virus.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study has been approved by the Ethics Committee of Jiangsu Provincial Center for Diseases Prevention and Control (approval No. JSJK2020-B007-02). Individual consent for this retrospective analysis was waived.

Detection reagents and instruments

The HIV1+2 antibody immunoblot kit produced by MP Biomedical (Singapore) was used to detect the above plasma samples. Identical testing kits and experimental procedures were adopted for both baseline and follow-up WB examinations. The MPAUTOBLOT20 protein immunoblotting instrument was used at room temperature for detection in strict accordance with the reagent instructions. The negative and positive controls included in the kit were used for each batch of tests. All WB tests were performed in strict accordance with the National Guideline for Detection of HIV/AIDS in China and the manufacturer’s instructions. Blinded testing was not performed, as the laboratory technicians were aware of the participants’ HIV status and clinical information during routine confirmatory testing. However, WB band interpretation was standardized and conducted independently by two senior laboratory technologists, with discrepancies resolved by consensus.

ART regimen

All enrolled participants received individualized ART regimens formulated by attending clinicians in accordance with the National Manual of Free Antiretroviral Treatment for AIDS.

WB diagnostic interpretation criteria

All visible bands were recorded, and results were interpreted according to the kit manufacturer’s recommended criteria, as follows:

• Negative: defined as either (I) no HIV-specific bands present; or (II) only the p17 antibody band detected, with no other HIV-specific bands.
• HIV-1-positive: required the presence of two envelope (ENV) bands (gp160/gp41 and gp120) plus at least one additional band from either the GAG group (p17, p24, or p55) or the POL group (p31, p51, or p66).
• HIV-1-positive with HIV-2 reactivity: met all criteria for HIV-1 positivity, with additional clear visibility of HIV-2-specific bands.
• Indeterminate: any HIV-specific band(s) present, but the pattern did not meet the criteria for HIV-1 positivity. Indeterminate results with visible HIV-2-specific bands were noted as “indeterminate with HIV-2 reactivity”.

All enrolled participants were confirmed HIV-1 positive at baseline by this set of uniform diagnostic standards.

Statistical analysis

This study adopted a retrospective registry-based design. All 401 enrolled patients had complete paired baseline and follow-up WB results, as well as complete demographic, clinical, and ART-related data. Therefore, no imputation or missing data handling was required. To justify the adequacy and statistical power of this cohort, the final sample size was evaluated based on the globally recognized events per variable (EPV) criterion for multivariable logistic regression. Variables entered into the multivariable logistic regression model were selected primarily based on previously published relevant literature. All clinically meaningful factors reported to be associated with HIV WB band changes in existing studies were included for analysis. For our primary endpoints involving prominent gag band dynamics, the absolute number of conversion events (e.g., 114 disappearance events for p55 and 66 for p17) significantly exceeded the minimum requirement of 10 events per independent covariate (EPV ≥10) for the 4 to 5 adjusted variables. This substantial event-to-variable ratio ensures robust statistical power, minimizes the risk of mathematical overfitting, and guarantees the stability of the calculated odds ratios. Excel was used for data entry and collation, and SPSS 26.0 software was used for statistical analysis. Enumeration data were expressed as frequencies and constituent ratios. Chi-squared test and Fisher’s exact test were used for intergroup comparisons. Multiple logistic regression was applied to calculate the odds ratios (ORs) and 95% confidence intervals (CIs) of major risk factors. Two-sided tests were used, and P<0.05 was considered statistically significant.

Results

General demographic characteristics

A total of 401 participants were included in this study, consisting of 352 males (87.78%) and 49 females (12.22%). The age distribution ranged from 16 to 73 years, with a mean age of 38.59±13.86 years. Among them, 155 cases (38.65%) were unmarried, 136 cases (33.92%) were married with spouses, and 110 cases (27.43%) were divorced or widowed. In terms of educational attainment, 49 cases (12.22%) had primary school education or below, 139 cases (34.66%) had junior high school education, 109 cases (27.18%) had senior high school or technical secondary school education, and 104 cases (25.94%) had junior college education and above. Transmission routes included 270 cases (67.33%) via same-sex sexual behavior and 131 cases (32.67%) via opposite-sex sexual behavior. In terms of ART duration, 79 cases (19.70%) had received treatment for 1 year, 101 cases (25.19%) for 3 years, 132 cases (32.92%) for 5 years, and 89 cases (22.19%) for 10 years.

For WB patterns, 118 cases (29.43%) showed initial full-band patterns, all meeting the criteria for HIV-1 antibody positive diagnosis. During follow-up, 79 cases (19.70%) presented full-band patterns, with 2 cases failing to meet the HIV-1 antibody positive diagnostic criteria. Two participants presented altered WB band profiles during follow-up. Case 1 exhibited baseline WB bands of gp160, gp120, p66, gp41, p31, p24 and p17; at follow-up, only gp160, gp120 and gp41 bands remained detectable. Case 2 had baseline bands including gp160, gp120, gp41 and p24, while follow-up WB only showed gp160, gp120 and gp41 bands.

Analysis of WB pattern changes

Analysis of the distribution and changes between initial and follow-up WB patterns showed that except for gp160 and gp120 bands, all other bands exhibited changes of “from present to absent” or “from absent to present”. Except for the p24 band, the total number of other bands decreased compared to the initial detection, as shown in Table 1.

Univariate analysis of band changes showed that only educational attainment influenced the change of p55 band, and different treatment durations influenced the change of p17 band, with P<0.05, as shown in Table 2. The changes of the remaining bands were not statistically significantly affected by the above factors.

Based on the univariate analysis results and other considered factors, further multivariate analysis showed that:

For the p55 band, individuals aged 40–49 years were more likely to exhibit band conversion from absent to present compared with those over 60 years, and those with senior high school or technical secondary school education were more prone to this change than those with college education or above.

For the p39 band, individuals with 1 year of treatment were less likely to experience band disappearance (from present to absent) than those with 10 years of treatment.

For the p17 band, unmarried individuals, those with 1 year of treatment, and those with 3 years of treatment were less likely to experience band disappearance than divorced/widowed individuals and those with 10 years of treatment, respectively. All results were statistically significant (P<0.05), as shown in Table 3.

The changes of the remaining bands were not statistically significantly affected by the above factors. To ensure full data transparency, the comprehensive univariate and multivariable logistic regression data for these non-significant bands (p66, p51, p41, p39, p31, and p24) are systematically provided in Tables S1,S2.

Analysis of influencing factors for WB band disappearance

Analysis of influencing factors for band changes from present to absent showed that the disappearance rates of p24 and p17 bands varied significantly across different treatment durations (P<0.05), and the disappearance rate of p24 bands also differed significantly across different age groups (P<0.05). The individual disappearance rates and corresponding statistical parameters for each of the evaluated WB bands across different treatment durations and age groups have been fully integrated into Tables 4,5.

Discussion

As HIV infection progresses, WB band patterns in infected individuals exhibit various dynamic changes. Typically, envelope protein antibody bands appear first, followed by other bands, with specific band patterns varying among individuals. Notably, along the development of the disease process, the p24 band may disappear in the late stage of infection (11,12). While several studies have reported uncertain WB patterns in untreated HIV-infected individuals (13-16), there have been no reports on WB band changes in patients receiving treatment. With the increasing degree of information, some infected individuals will use drugs irregularly by themselves, which brings challenges to the confirmation of HIV antibodies. Therefore, it is very important to understand the change rules of WB bands after drug use.

In this report, the band pattern distribution characteristics at the initial confirmation were similar to those in other literatures (17), yet no comparable findings regarding their longitudinal dynamics have been retrieved from previous publications. After treatment, three types of changes were observed in WB bands, namely appearance from absence, no change, and disappearance. Among these, the envelope protein antibody bands encoded by the env gene showed the highest stability, with only 2.2% (9/401) of the gp41 bands showing changes The pol gene bands were the next most stable, whereas bands in the gag region (except p24) showed the highest rates of change. The analysis results of the change factors of each band showed that only the bands in the gag region could be affected by different factors, leading to band changes. The bands in the gag region belong to core proteins, which play a role in stabilizing and promoting viral entry into the nucleus. Research shows that highly immunogenic env glycoproteins induce long-lived plasma cells in bone marrow; these cells secrete anti-env antibodies for decades without viral antigen stimulation, leading to stable env WB bands during follow-up (10). But anti-gag antibodies targeting internal viral core proteins rely on sustained viral replication for persistent antigen supply (14). We speculate that the high variability of these bands is potentially associated with the intrinsic stability of core proteins, which requires further verification (18).

In clinical practice, the disappearance of WB bands can complicate HIV antibody confirmation and result interpretation. Therefore, we pay more attention to the factors influencing band disappearance. Our findings showed that the p39 band was more likely to disappear in individuals who had been treated for ten years compared to those treated for one year; for the p17 band, higher disappearance rates were observed among divorced or widowed individuals and those treated for ten years, compared to unmarried individuals and those with shorter treatment durations; Similarly, p24 band disappearance, was more frequent among individuals aged 50–59 years and those treated for ten years. Overall, the longer the treatment time and the older the age, the more likely the bands are to disappear. This phenomenon may be partially explained by disease progression or age-related decline in immune function, which is a reasonable hypothesis rather than a confirmed conclusion (19). Consequently, it is essential to closely monitor patients receiving long-term treatment or elderly patients, timely evaluation of potential drug resistance, and adjust treatment regimens accordingly. The higher rate of band disappearance in divorced individuals, compared to unmarried individuals, may also be partially explained by the older average age in the divorced group.

In this report, two cases showed a transition from positive band patterns to indeterminate ones. For the first case, it may be related to the weak immune response of the body during the late stage of the disease at the initial detection. Antibodies appeared after hospitalization, but subsequently, due to factors such as drug insensitivity or poor compliance, the body’s immune system was damaged again, resulting in no response of certain specific bands (such as p24). This finding is similar with our research before (7) and other researches (8). As for the other case, they may be associated with earlier initiation of ART. After taking the medications, the viral load decreased, and insufficient antigen stimulation in the body led to a weakened antibody response (1,7). Our findings are similar with other research, studies of early-treated HIV-infected infants and older individuals have demonstrated that when patients receive cART shortly after initial HIV infection, HIV antibodies might not develop (20-22). Nearly one-third of perinatally infected children had a negative or indeterminate WB result later in life (1). These two special cases remind us that we should be cautious in diagnosing the conditions of patients in the critical stage of infection, those seeking pre-exposure prophylaxis consultation, and late stage patients with symptoms. We should enhance the popularization of nucleic acid testing (NAT) among the above-mentioned populations, closely follow up the progression of their WB bands, and carefully address the challenges posed by pre- (PrEP) and post-exposure prophylaxis (PEP) to laboratory testing and the diagnosis of HIV infection.

Collectively, this study offers several meaningful academic and clinical implications for HIV diagnostic testing and population surveillance. First, in terms of research methodology, this work complements existing static, single-timepoint cross-sectional evidence via an 11-year longitudinal cohort framework, which enables characterization of intra-individual dynamic changes in WB band profiles over time. Second, the present analysis adds novel epidemiological evidence regarding distinct antibody waning patterns among products encoded by different HIV gene regions following ART initiation. The data indicate that gag-region bands (p55, p39, p17) appear more susceptible to fading with extended treatment duration and advancing age. Lastly, from a public health perspective, these results provide reference for interpreting atypical, diagnostically discordant HIV specimens amid growing uptake of PrEP, PEP and unsupervised antiretroviral use. The observation of attenuated serologic reactivity, including borderline or WB-negative profiles after long-term therapy, suggests that supplementary NAT should be considered for cases with inconsistent serological results, which may help reduce the risk of missed diagnosis.

This study has several limitations. As a multicenter retrospective study, complete baseline clinical data were unavailable for some participants. Specifically, baseline viral load was not routinely tested at initial HIV confirmation, making relevant confounders uncollectible and unadjusted in statistical analyses.

Additionally, other logistical elements, including non-fixed follow-up intervals and multi-site sample transport variances, should be acknowledged merely as potential residual confounding factors rather than systematic biases. Furthermore, WB used in this research is inherently a semi-quantitative detection method. Manual visual interpretation of WB bands inevitably introduces subjective variability in band judgment and intensity evaluation, which may exert subtle influences on the consistency of band classification results. Nonetheless, our strict reliance on complete, paired data across all participating sites ensures that the observed longitudinal trends in gag band dynamics remain internally valid and representative.

Conclusions

To sum up, changes in the gag region bands p55, p39, and p17 were associated with treatment duration, marital status and age. Specifically, longer treatment duration and advanced age increased the risk of WB band disappearance. Accordingly, for elderly individuals or those with irregular self-medication presenting with indeterminate HIV antibody WB results, NAT is recommended as a supplementary assay, and regular follow-up is required to reduce HIV transmission risks. Collectively, our findings highlight the need for cautious interpretation of WB results among treated patients and support the use of NAT for confirmation in cases with inconsistent serological findings.

Acknowledgments

The authors would like to take this opportunity to thank the Centers for Disease Control and Prevention of Yangzhou, Suzhou, and Wuxi for their support of this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jphe.amegroups.com/article/view/10.21037/jphe-2026-0020/rc

Data Sharing Statement: Available at https://jphe.amegroups.com/article/view/10.21037/jphe-2026-0020/dss

Peer Review File: Available at https://jphe.amegroups.com/article/view/10.21037/jphe-2026-0020/prf

Funding: This work was supported by the Jiangsu Province Medical Key Discipline (“14th Five-Year Plan” Epidemiology) (No. ZDXK202250).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jphe.amegroups.com/article/view/10.21037/jphe-2026-0020/coif). All authors report this work was supported by the Jiangsu Province Medical Key Discipline (“14th Five-Year Plan” Epidemiology) (No. ZDXK202250). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study has been approved by the Ethics Committee of Jiangsu Provincial Center for Diseases Prevention and Control (approval No. JSJK2020-B007-02). Individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Fuller T, Kerin T, Cortado R, et al. Sustained Virologic Suppression Reduces HIV-1 DNA Proviral Levels and HIV Antibodies in Perinatally HIV-Infected Children Followed from Birth. Viruses 2022;14:2350. [Crossref] [PubMed]
2. Liu X, Wu C, Yuan Y, et al. Clinical performance evaluation of an HIV Duo assay: From HIV screening to acute and non-acute HIV infection detection. Clin Chim Acta 2025;565:119949. [Crossref] [PubMed]
3. Liu P, Tang L, Kong WH, et al. Anti-HIV-1 antibodies based confirmatory results in Wuhan, China, 2012-2018. PLoS One 2020;15:e0238282. [Crossref] [PubMed]
4. Wang W, Meng Z, Yang R, et al. Analysis of common patterns of HIV antibody bands detected by immunoblot and their value in follow-up. Lab Med 2020;35:95-9.
5. Cao M, Yang DZ, Wu ZL, et al. Analysis of HIV antibody confirmatory results in Ningxia from 2017 to 2019. Ningxia Med J 2021;43:337-40.
6. Xiao Q, Xing WG, Xue XM, et al. Distribution of Western blot bands in HIV-infected individuals. Chin J AIDS STD 2012;18:195-6.
7. Zhang Z, Lu J, Xu XQ, et al. Analysis of Western blot bands and viral load in individuals with indeterminate and negative HIV-1 antibodies. J Med Postgraduates 2022;35:131-5.
8. Ouyang L, Zhang TJ, Zhang YH, et al. Characteristics of Western blot band patterns of HIV-1 antibody-positive individuals in different clinical stages. Fudan Univ J Med Sci 2017;(5):664-7.
9. Bian SC, Yu G, Ji KCN, et al. Study on the use of Western blot antibody bands for determining new HIV infections. Chin J AIDS STD 2022;28:191-4.
10. Guo C, Ji LY, Wu YB, et al. Analysis of protein immunoblot patterns in blood samples from HIV-1 infected/AIDS patients in different clinical stages. Lab Med 2015;30:145-8.
11. Lin GH, Fang GZ. Relationship between Western blot band types and immune suppression in newly diagnosed HIV-infected individuals in Jieyang. Henan Med Res 2021;30:4269-71.
12. Xie SY, Chen XJ. Analysis of Western blot band patterns in 917 HIV antibody positive and indeterminate cases. Chin J AIDS STD 2018;24:1189-91.
13. Liu JF, Zhang M, Sun AS, et al. Demographic and Western blot band characteristics of 41 pairs of HIV-1 seroconcordant couples in Yunnan. Chin J Prev Med 2020;21:60-4.
14. Lara-Aguilar V, Llamas-Adán M, Brochado-Kith Ó, et al. Low-level HIV-1 viremia affects T-cell activation and senescence in long-term treated adults in the INSTI era. J Biomed Sci 2024;31:80. [Crossref] [PubMed]
15. Tang Y, Zhong P, Zhu LY. Analysis of a case of incomplete bands in HIV antibody immunoblot assay. Chin J AIDS STD 2014;20:495-6.
16. Zou JB, Sun TY, Li X, et al. Laboratory report of a pregnant woman with incomplete bands in HIV-1 protein immunoblot assay. Chin J AIDS STD 2017;23:354-6.
17. Wei W. Analysis of Western blot patterns and initial CD4+ T lymphocyte counts in HIV-1 antibody positive individuals in Liaoyang, 2015–2016. Chin J Health Lab Technol 2018;28:619-21.
18. Zhao S, Wang W, Li S, et al. The prevalence of low-level viraemia and its association with virological failure in people living with HIV: a systematic review and meta-analysis. Emerg Microbes Infect 2025;14:2447613. [Crossref] [PubMed]
19. Lao X, Zhang H, Deng M, et al. Incidence of low-level viremia and its impact on virologic failure among people living with HIV who started an integrase strand transfer inhibitors: a longitudinal cohort study. BMC Infect Dis 2024;24:8. [Crossref] [PubMed]
20. Persaud D, Patel K, Karalius B, et al. Influence of age at virologic control on peripheral blood human immunodeficiency virus reservoir size and serostatus in perinatally infected adolescents. JAMA Pediatr 2014;168:1138-46. [Crossref] [PubMed]
21. Foster C, Domínguez-Rodríguez S, Tagarro A, et al. The CARMA Study: Early Infant Antiretroviral Therapy-Timing Impacts on Total HIV-1 DNA Quantitation 12 Years Later. J Pediatric Infect Dis Soc 2021;10:295-301. [Crossref] [PubMed]
22. Bitnun A, Samson L, Chun TW, et al. Early initiation of combination antiretroviral therapy in HIV-1-infected newborns can achieve sustained virologic suppression with low frequency of CD4+ T cells carrying HIV in peripheral blood. Clin Infect Dis 2014;59:1012-9. [Crossref] [PubMed]