BGPT: Paper Review: Association of HLA-DQ and -DR Alleles with Protection from or Infection with HIV-1

Explore by Goal

Quick Explanation Copied

Paper-at-a-glance (skeptical, evidence-weighted)

Study type: small human case-control genetic association (HLA-DQ/DR) with ethnic subgrouping (African American vs Caucasian).
Key reported signals: DQB1*0603 associated with HIV-1 infection in Caucasians (reported RR ~4.67; P~0.0335), but not in African Americans.
Reported protection: DQB1*03032 described as protective in African Americans and also protective in Caucasians (with subgroup P values reported).
Central skepticism: multiple HLA alleles tested in small strata without a clear multiple-testing correction strategy for the overall number of comparisons, increasing false-positive risk.

Go deeper

Long Explanation

Association of HLA-DQ and -DR Alleles with Protection from or Infection with HIV-1

Visual, skeptical paper review (host genetics; HIV-1 infection vs uninfected controls; ethnic stratification)

DOI: 10.1006/exmp.1999.2287

Known biological context (what HLA class II can plausibly affect)

HLA class II molecules (including HLA-DQ and HLA-DR) present peptide antigens to CD4+ T cells; thus, class II variation can plausibly influence adaptive immune response quality, which can affect infection susceptibility and control.
Mechanistically, HLA class II effects are often allele- and antigen-specific and can be modified by population differences (e.g., linkage disequilibrium structure, allele frequencies, and HIV diversity), so genetic associations may differ across ancestries.

Figure 1. Reported allele frequency differences used in the main claims (from tables in the paper)

How to read Figure 1

Frequencies plotted are allele prevalence percentages as presented in the paper text/tables for the specified contrasts (HIV+ vs HIV−), with ethnicity-specific patterns reported for DQB1*0603 and DQB1*03032.

Figure 2. Effect sizes (odds ratios / relative risks) as reported (selected significant or highlighted alleles)

Skeptical note

The plotted effect sizes are those explicitly reported in the manuscript’s tables; however, the overall interpretation is limited by small sample sizes and the large number of alleles tested within strata, which can inflate spurious signals if correction for multiple comparisons is not comprehensive.

1) Study design & sample (what was actually measured)

Design: case-control genetic association comparing HIV-1 positive adults vs HIV-1 negative controls, with subgroup analyses by race/ethnicity (African American vs Caucasian).
Genotyping method: PCR amplification of HLA-DQB1 and HLA-DRB1 regions followed by reverse dot-blot hybridization using LiPA/INNO-LiPA (Murex Diagnostics) and LiPA software.
Controls: 47 HIV-1 negative “healthy controls” were included; the excerpt does not provide details on exposure history for these controls (a key interpretability point: uninfected ≠ unexposed).

2) Primary claims & how well the evidence supports them

2.1 DQB1*0603 as infection-associated (Caucasians)

The manuscript reports DQB1*0603 as positively associated with HIV-1 infection in Caucasians (with an odds ratio and P value reported in the tables).
Uncertainty: the same study repeatedly emphasizes allele frequency comparisons across relatively small ethnic subgroups; with multiple alleles tested, one significant-ish result may reflect random variation rather than a robust genetic effect.

2.2 DQB1*03032 as protective (both ethnic groups)

The paper reports DQB1*03032 frequencies higher among HIV− than HIV+ individuals within both African Americans and Caucasians, consistent with “protection from infection.”
Interpretation risk: without clarifying whether HIV− controls are exposed seronegative vs unexposed, “protection” can be confounded by differential exposure risk rather than biological resistance.

2.3 DRB1*04 as infection-associated (Caucasians)

The paper reports HLA-DRB1*04 as positively associated with HIV-1 infection in Caucasians (with P≈0.0060 and strong frequency contrast reported between HIV+ and African American HIV+).

3) Statistical validity: what could mislead

3.1 Multiple testing across many alleles

The study reports many DQB1 and DRB1 allele frequencies and tests multiple comparisons across ethnic strata. Without explicit, comprehensive correction across the full number of alleles tested, the probability of at least one false-positive “significant” result increases.

3.2 Sample size / allele rarity

Allele-level comparisons can have very small counts (including zeros and “ND” where OR is not defined), making estimates unstable and inflating uncertainty.

3.3 Population stratification / linkage disequilibrium

HLA loci have strong linkage disequilibrium and are ethnically structured. Reported “ethnic-specific” effects can arise from linkage with other causal variants or from differences in allele/haplotype backgrounds.

4) Biological plausibility: what mechanisms could connect these associations to HIV-1

In humans, certain DRB1 alleles are linked to differences in breadth of HIV-specific CD4+ T cell responses and viral control, consistent with class II influencing antigen presentation and helper T cell quality.
The paper’s specific “DQ/DR allele → infection susceptibility” framing is plausible, but the presented data are association-only and do not include functional peptide-binding / CD4 response assays for the alleles claimed.

5) What would most likely change (or falsify) these conclusions?

Replication in independent cohorts with adequate power per allele and explicit multiple-testing correction.
Exposure characterization: inclusion of exposed-seronegative controls would help separate biological “resistance” from exposure-risk differences.
Functional confirmation: demonstration that the implicated DQ/DR alleles present specific HIV epitopes that measurably change CD4 specificity/breadth/avidity and map onto viral control phenotypes.

6) Directed evidence connections (graphical knowledge map)

The causal arrow map is grounded in the paper’s association design and in later DRB1 functional evidence connecting class II to CD4 response properties and viral control .

7) Practical takeaway (with confidence grading)

Main takeaway: this late-1990s case-control study reports ethnicity-stratified associations between certain HLA-DQ/DR alleles and HIV-1 infection vs lack of infection (and mentions allele-associated progression phenotypes), but the evidence strength is limited by small sample sizes and likely multiple-testing burden.
Confidence (in the existence of a true biological effect at these specific alleles): low to moderate without replication + functional assays for these particular DQ/DR alleles.

Author review links

Feedback:

Updated: March 23, 2026