An inflammation-related gene result that did not replicate
Lack of Association Between MIR155 Gene Variant rs767649 and Risk for Parkinson's Disease.
AI-authored reading companion. Supporting quotations match the source text; this does not verify the interpretation. No expert review yet.
Hortensia Alonso-Navarro, Sofía Ladera-Navarro, Pedro Ayuso, Pau Pastor, Ignacio Álvarez, Miquel Aguilar, Elena García-Martín, José A G Agúndez, Félix Javier Jiménez-Jiménez · International journal of molecular sciences · 2026
Adapted from the original publication. License: https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.
Spanish researchers tested a genetic link to Parkinson’s previously reported in China and did not find convincing evidence for it in their participants. The genetic version they studied was uncommon, so a small effect could still have been missed. This result leaves the proposed link uncertain.
Read the supporting text
In conclusion, after considering all the limitations of the study, our results do not support a significant association between the MIR155 rs767649 variant and PD risk or age at disease onset in this Spanish Caucasian cohort. However, given the low statistical power of the study, these results are inconclusive rather than definitively negative; therefore, a modest genetic effect cannot be excluded, and larger studies in independent populations are warranted to further clarify the possible role of this variant in PD susceptibility.
In this study, 459 patients diagnosed with PD participated according to the UK Parkinson’s Disease Society Brain Bank Clinical Diagnostic Criteria [40], along with 460 healthy controls matched by age and sex. Both PD patients and controls were over 18 years of age.
In the current study, we failed to replicate the association of the MIR155 gene variant rs767649 with PD reported in a previous study [20], using a larger sample size. We found no association between this variant and PD risk, either in the overall sample or after stratification by sex. We also found no influence of the MIR155 rs767649 variant on the age at PD onset.
Finally, the low frequency of the rs767649 A allele, particularly the very small number of AA homozygotes (one control and two PD patients), resulted in wide confidence intervals and limited statistical power, reducing the precision of the estimated odds ratios and limiting our ability to detect modest genetic effects.
In brief
A previous Chinese study linked a DNA variant near MIR155, a gene involved in regulating inflammation, to Parkinson’s risk. Could a larger Spanish study reproduce that association?
Read the supporting text
A recent case–control association study conducted in the Chinese Han population, involving 110 patients with PD and 120 healthy controls, showed a decreased risk of PD in subjects carrying the MIR155 rs767649 A allele, as well as greater disease severity, poorer scores on a recognized cognitive scale, and higher serum levels of miR-155 in PD patients carrying the MIR155 rs767649 TT genotype [20]. The present study aimed to replicate the potential role of rs767649 variants in PD risk in a larger cohort of Spanish Caucasian patients with PD and healthy controls.
The researchers compared 459 people with Parkinson’s and 460 healthy controls matched by age and sex. This was a genetic case–control comparison, not a trial of an inflammation treatment.
Read the supporting text
In this study, 459 patients diagnosed with PD participated according to the UK Parkinson’s Disease Society Brain Bank Clinical Diagnostic Criteria [40], along with 460 healthy controls matched by age and sex. Both PD patients and controls were over 18 years of age.
They found no statistically significant association with Parkinson’s or age at onset. That result failed to reproduce the earlier association; it does not show that inflammation itself is irrelevant.
Read the supporting text
In the current study, we failed to replicate the association of the MIR155 gene variant rs767649 with PD reported in a previous study [20], using a larger sample size. We found no association between this variant and PD risk, either in the overall sample or after stratification by sex. We also found no influence of the MIR155 rs767649 variant on the age at PD onset.
The uncommon A version appeared in only about 5% of gene copies. Just two patients and one control had two A copies. That leaves considerable uncertainty and makes small effects difficult to detect, even in a study of 919 people.
Read the supporting text
Finally, the low frequency of the rs767649 A allele, particularly the very small number of AA homozygotes (one control and two PD patients), resulted in wide confidence intervals and limited statistical power, reducing the precision of the estimated odds ratios and limiting our ability to detect modest genetic effects.
Keep this as an inconclusive replication result, alongside the earlier positive study. It narrows confidence in a particular genetic claim without disproving the wider inflammation hypothesis. Larger independent studies are needed.
Read the supporting text
In conclusion, after considering all the limitations of the study, our results do not support a significant association between the MIR155 rs767649 variant and PD risk or age at disease onset in this Spanish Caucasian cohort. However, given the low statistical power of the study, these results are inconclusive rather than definitively negative; therefore, a modest genetic effect cannot be excluded, and larger studies in independent populations are warranted to further clarify the possible role of this variant in PD susceptibility.
When was the research done?
Participants were recruited between 2003 and 2018. The paper was published in 2026. The recruitment period tells us when people entered the underlying study; it does not tell us the exact date of every DNA test.
Read the supporting text
Recruitment of both PD patients and controls was carried out between 2003 and 2018.
Where did it happen?
Patients came from movement-disorder units at four hospitals. The paper identifies controls from Hospital Universitario Infanta Cristina in Badajoz and Clínica Universitaria de Navarra in Pamplona, Spain. The study describes the participants as Spanish people of European ancestry.
Read the supporting text
PD patients were recruited from the Movement Disorders Units of four hospitals and were examined before inclusion by clinical neurologists with extensive experience in the diagnosis of Parkinson’s disease. Controls were recruited from Hospital Universitario Infanta Cristina in Badajoz, Spain (most of them staff or students of the hospital) and from Clínica Universitaria de Navarra in Pamplona, Spain (spouses of PD patients).
The full story
A small genetic clue within a bigger idea
The paper starts from a plausible idea: inflammation may contribute to the processes that damage nerve cells in Parkinson’s, and a molecule called miR-155 helps regulate inflammatory responses. MicroRNAs are short RNA molecules that influence how other genes are used. The MIR155 gene provides instructions for making miR-155. This study did not test every part of that biological pathway. It asked a narrower question: is one particular DNA difference, called rs767649, associated with having Parkinson’s? Keeping the pathway and this individual genetic marker separate is essential to understanding the result.
Read the supporting text
MicroRNAs (miRNAs) are a class of endogenous non-coding RNAs of approximately 21–25 nucleotides that occur naturally and are involved in the regulation of gene expression. Cancer, immune dysfunction, and some neurological diseases are associated with aberrant expression of these molecules. Specifically, the multifunctional miRNA miR-155 is implicated in processes such as innate [6,7] and adaptive immunity [8,9], hematopoiesis [7], cell proliferation and apoptosis [10], host defense [9], and inflammation [11,12,13,14,15]. miR-155 is upregulated by cytokines and pathogens and modulates the production of tumour necrosis factor alpha (TNF-α), interleukin-6 (IL-6), interleukin-1β (IL-1β), interferon-γ (IFN-γ), and interferon regulatory factor 3 (IRF3) [13,14].
miR-155 is encoded by the microRNA-155 gene (MIR155, also named MIRN155, miRNA 155 or mir-155; chromosome 21q21.3; gene ID 406947; MIM 609337; https://www.ncbi.nlm.nih.gov/gene/406947, accessed on 8 April 2026). One of the most common variants in the MIR155 gene, the rs767649 single nucleotide polymorphism (SNP), is a T > A polymorphism that locates at the regulatory/promoter region of the gene and can affect miR-155 expression levels (with the TT genotype often associated with higher miR-155 expression levels) [16].
A prior study in a Chinese Han population compared 110 people with Parkinson’s and 120 controls. It reported that people carrying the A version of rs767649 had lower Parkinson’s risk. The Spanish team attempted a replication: checking whether a result appears again in a different group. Repetition is particularly important for genetic associations, because a finding in one relatively small sample can depend on chance, the surrounding genetic background or how participants were selected. The new study used a larger overall sample, but being larger does not automatically make it decisive when the version of interest is uncommon.
Read the supporting text
A recent case–control association study conducted in the Chinese Han population, involving 110 patients with PD and 120 healthy controls, showed a decreased risk of PD in subjects carrying the MIR155 rs767649 A allele, as well as greater disease severity, poorer scores on a recognized cognitive scale, and higher serum levels of miR-155 in PD patients carrying the MIR155 rs767649 TT genotype [20]. The present study aimed to replicate the potential role of rs767649 variants in PD risk in a larger cohort of Spanish Caucasian patients with PD and healthy controls.
Finally, the low frequency of the rs767649 A allele, particularly the very small number of AA homozygotes (one control and two PD patients), resulted in wide confidence intervals and limited statistical power, reducing the precision of the estimated odds ratios and limiting our ability to detect modest genetic effects.
Who took part, and what did the team test?
There were 459 patients and 460 controls, all adults who self-reported Spanish European ancestry. Patients were recruited from movement-disorder units at four hospitals and examined by experienced neurologists. Controls included hospital staff or students and spouses of patients. Recruitment took place between 2003 and 2018, and these participants had also contributed to earlier genetic studies from the group. Age and sex were matched to reduce obvious differences between cases and controls. This was not a random sample of everyone living in Spain, and the investigators did not have detailed genetic ancestry markers to check subtler population differences.
Read the supporting text
In this study, 459 patients diagnosed with PD participated according to the UK Parkinson’s Disease Society Brain Bank Clinical Diagnostic Criteria [40], along with 460 healthy controls matched by age and sex. Both PD patients and controls were over 18 years of age.
PD patients were recruited from the Movement Disorders Units of four hospitals and were examined before inclusion by clinical neurologists with extensive experience in the diagnosis of Parkinson’s disease. Controls were recruited from Hospital Universitario Infanta Cristina in Badajoz, Spain (most of them staff or students of the hospital) and from Clínica Universitaria de Navarra in Pamplona, Spain (spouses of PD patients).
None of the healthy controls included in the study had a family history of PD, other movement disorders, or other neurological diseases. Recruitment of both PD patients and controls was carried out between 2003 and 2018.
The demographic data of both PD patients and control subjects, who have participated in previous case–control association studies [41,42], are summarized in Table 4. All participants were self-reported Spanish Caucasians of European ancestry. Because detailed ancestry-informative markers were not available, formal ancestry-stratified analyses could not be performed. We agree that recruitment of some controls among spouses of PD patients may represent a potential source of selection bias, and we have acknowledged this limitation in Section 3.
At the DNA position studied, the common version is represented by T and the less common version by A. A genotype describes the pair of versions a person carries: TT, AT or AA. An allele is one of those individual versions. The team used a laboratory DNA test called a TaqMan assay on DNA from blood cells. They then compared genotype and allele frequencies between patients and controls. They also analyzed several inheritance patterns: for example, whether carrying any A copy, or specifically two A copies, appeared related to disease. These are alternative ways to test the same marker, not separate discoveries.
Read the supporting text
miR-155 is encoded by the microRNA-155 gene (MIR155, also named MIRN155, miRNA 155 or mir-155; chromosome 21q21.3; gene ID 406947; MIM 609337; https://www.ncbi.nlm.nih.gov/gene/406947, accessed on 8 April 2026). One of the most common variants in the MIR155 gene, the rs767649 single nucleotide polymorphism (SNP), is a T > A polymorphism that locates at the regulatory/promoter region of the gene and can affect miR-155 expression levels (with the TT genotype often associated with higher miR-155 expression levels) [16].
Genotyping of MIR155 rs767649 was performed by using real-time PCR (Applied Biosystems 7500 qPCR thermocycler, Waltham, MA, USA) with specific custom-designed TaqMan assay (C___2212229_10, Life Technologies, Alcobendas, Madrid, Spain) on genomic DNA obtained from peripheral leukocytes of venous blood samples of patients diagnosed with PD and controls.
The MIR155 rs767649 allelic frequencies were in Hardy–Weinberg equilibrium in both study groups, with p-values of 0.406 and 0.996, respectively, in PD patients and controls. The results of genotypes in PD patients and control subjects, according to codominant, dominant, recessive, and additive models, are summarized in Table 1, and those of allelic variants in Table 2. Under the codominant model, odds ratios were calculated separately for the comparisons AT versus TT and AA versus TT, using TT as the reference genotype. The extremely small number of AA homozygotes prevented a reliable evaluation of recessive effects.
What they found
Among controls, 418 people were TT, 41 were AT and one was AA. Among patients, 415 were TT, 42 were AT and two were AA. In other words, the distributions were very similar. At the gene-copy level, A accounted for 4.7% of control copies and 5.0% of patient copies. The reported overall allele odds ratio was 1.076, with a 95% confidence interval from 0.70 to 1.65. A value of 1 means equal odds in the comparison. This wide interval includes both lower and higher odds, so the estimate does not reliably establish either protection or increased susceptibility.
Read the supporting text
Control | 418 (90.87%) | 41 (8.91%) | 1 (0.22%) | | | | | | | |
PD | 415 (90.41%) | 42 (9.15%) | 2 (0.44%) | 1.03 (0.66–1.62)2.01 (0.18–22.30) | 0.790, 0.910 | 1.06 (0.68–1.65) | 0.820, 0.910 | 2.01 (0.18–22.23) | 0.620, 0.910 | 1.03 (0.66–1.62) | 0.910, 0.910
Control | 877 (95.3%) | 43 (4.7%) | | |
PD | 872 (95.0%) | 46 (5.0%) | 1.076 (0.70–1.65) | 0.737 | 0.50 (0.50–0.51)
The authors found no statistically significant association across their genotype comparisons, including separate analyses of men and women. Adjusting for age and sex did not change the overall conclusion. They also found no statistically significant difference in the age when Parkinson’s began or in disease duration across genotype groups. With only two AA patients, an apparent difference in an average age would be especially unstable. The analysis also accounted for multiple comparisons, which matters because trying many statistical comparisons can produce apparently interesting results by chance. None of this turns a non-significant result into proof that the true effect is exactly zero.
Read the supporting text
No significant differences were found under any of the four genetic models in the frequency of the different genotypes between PD patients and healthy controls, either in the overall sample or after stratification by sex (Table 1). Similarly, the allele frequencies of the two variants did not differ significantly between the two groups, either in the overall analysis or after stratification by sex (Table 2). Finally, the age at PD onset and the evolution time of PD did not differ significantly among the three possible genotypes of the MIR155 rs767649 variant (Table 3). To address potential residual confounding, a multivariate logistic regression analysis adjusted for age and sex was conducted. The adjusted results were consistent with those obtained in the univariate analyses and did not reveal any statistically significant association between rs767649 and Parkinson’s disease susceptibility.
Intergroup comparisons were performed using the χ2 test or Fisher’s exact test, as appropriate, and correction for multiple testing was performed using the false discovery rate procedure [44]. For each intergroup comparison, including all genotypes or all alleles, both crude and corrected values (p and Pc, respectively) were calculated.
Finally, the low frequency of the rs767649 A allele, particularly the very small number of AA homozygotes (one control and two PD patients), resulted in wide confidence intervals and limited statistical power, reducing the precision of the estimated odds ratios and limiting our ability to detect modest genetic effects.
Why the answer is still uncertain
Statistical power describes a study’s ability to detect an effect of a given size. Here, the rare allele and the tiny AA subgroup limit what can be learned. Having many participants who all carry the common version supplies less information about a rare variant than the total sample size might suggest. The authors explicitly call their result inconclusive rather than definitively negative, and acknowledge the possibility of a type II error: missing a real association. The fair plain-language conclusion is that this study did not find convincing evidence of the proposed association, while modest effects remain possible.
Read the supporting text
Finally, the low frequency of the rs767649 A allele, particularly the very small number of AA homozygotes (one control and two PD patients), resulted in wide confidence intervals and limited statistical power, reducing the precision of the estimated odds ratios and limiting our ability to detect modest genetic effects. Consequently, although no statistically significant association between rs767649 and PD risk was identified, a type II error cannot be excluded, and our findings should be interpreted as showing no evidence of association rather than definitively excluding a contribution of this variant to PD susceptibility. Substantially larger cohorts, as well as independent replication studies in other populations, will be required to more precisely evaluate the potential role of rs767649 in Parkinson’s disease and to adequately assess the small effect sizes that may be associated with this variant.
In conclusion, after considering all the limitations of the study, our results do not support a significant association between the MIR155 rs767649 variant and PD risk or age at disease onset in this Spanish Caucasian cohort. However, given the low statistical power of the study, these results are inconclusive rather than definitively negative; therefore, a modest genetic effect cannot be excluded, and larger studies in independent populations are warranted to further clarify the possible role of this variant in PD susceptibility.
The paper suggests that differences in genetic background could contribute to the disagreement with the Chinese study. Sometimes a tested marker is not the biologically active change itself, but tends to be inherited alongside another nearby change. That relationship can differ between populations. Interactions with other genes could also differ. These are proposed explanations, not findings this study established. Recruitment and chance remain relevant too. Our library should link the two reports as a replication pair while preserving their population descriptions, so a search does not silently combine them as though they studied identical circumstances.
Read the supporting text
The discrepancy between the results obtained in the Chinese Han population and those observed in our Spanish Caucasian cohort may be related to population-specific genetic factors. Parkinson’s disease is a complex disorder in which the effect of individual genetic variants may depend on interactions with other susceptibility loci whose frequencies differ among populations. In addition, rs767649 itself may not be the causal variant but rather a marker in linkage disequilibrium with a nearby functional variant. Differences in linkage disequilibrium patterns between ethnic groups could therefore lead to the detection of an association in one population but not in another. These possibilities should be considered when interpreting the inconsistent findings across studies and warrant further investigation in larger cohorts from diverse ancestral backgrounds.
Where the research can go next
The investigators did not measure how much miR-155 each participant produced according to genotype, because suitable samples were unavailable for everyone and the AA group was too small. They did not test an anti-inflammatory medicine or measure whether changing miR-155 slows Parkinson’s. Therefore this paper cannot settle whether miR-155 is a useful treatment target. Its contribution is to challenge the certainty of one proposed risk-marker association and point toward larger, independent studies that combine DNA measurements with functional measurements. Including such results prevents a discovery dataset from becoming a collection of positive claims only.
Read the supporting text
Third, we did not perform functional analyses evaluating miR-155 expression according to rs767649 genotype. Although such analyses could have provided valuable biological support for the genetic findings, suitable samples were not available for all participants, and the extremely small number of AA homozygotes in our cohort would have limited the interpretability of genotype-expression comparisons. Future studies integrating genotyping with measurements of miR-155 expression in serum, plasma, or PBMCs may help clarify the functional significance of this variant.
In conclusion, after considering all the limitations of the study, our results do not support a significant association between the MIR155 rs767649 variant and PD risk or age at disease onset in this Spanish Caucasian cohort. However, given the low statistical power of the study, these results are inconclusive rather than definitively negative; therefore, a modest genetic effect cannot be excluded, and larger studies in independent populations are warranted to further clarify the possible role of this variant in PD susceptibility.
Is this research continuing?
We have not yet checked for a later publication, registered follow-up study or update from this team. The next steps below are proposals in this paper, not confirmation that the work is underway.
Based on the full acquired article text, including available tables and figure captions. Figure images, raw data and separate supplementary files were not independently examined. This is an explanatory adaptation, not a line-by-line translation.