The article entitled “The Clinical Course of Patients with Preschool Manifestation of Type 1 Diabetes is Independent of the HLA DR-DQ Genotype” contained a calculation error in Table 2 and the statistical methods used were not completely described. The additional information (in italics) and corrected Table 2 are given below:
Table 2.
Analysis of the DRB1-DQA1-DQB1 second alleles in heterozygous DRB1*03/x (DR3-DQ2) and HLA DRB1*04/x (DR4-DQ8) patient and controls, carrying one high-risk allele. Data were separately analysed for HLA DRB1*03/x (DR3-DQ2) and DRB1*04/x (DR4/DQ8) heterozygous groups.
| Patients with DRB1*03/x | ||||||
| DRB1* | DQA1* | DQB1* | Controls | T1D Patients | Odds Ratio (CI) | p |
| n (%) | n (%) | |||||
| 01:01 | 01:01 | 05:01 | 359 (10.5) | 11 (29.7) | 3.62 (1.76; 7.39) | <0.001 |
| 07:01 | 02:01 | 02:02 | 332 (9.7) | 2 (5.4) | 0.53 (0.13; 1.98) | 0.575 |
| 08:01 | 04:01/02 | 04:02 | 105 (3.1) | 5 (13.5) | 4.95 (2.05; 12.22) | 0.006 |
| 09:01 | 03:02 | 03:03 | 23 (0.7) | 4 (10.8) | 17.95 (6.37; 53.99) | <0.001 |
| 12:01 | 05:05 | 03:01 | 94 (2.7) | 2 (5.4) | 2.03 (0.47; 7.78) | 0.274 |
| 13:01 | 01:03 | 06:03 | 281 (8.2) | 2 (5.4) | 0.64 (0.15; 2.39) | 0.765 |
| 13:02 | 01:02 | 06:04 | 142 (4.1) | 3 (8.1) | 2.04 (0.65; 6.03) | 0.200 |
| 16:01 | 01:02 | 05:02 | 120 (3.5) | 8 (21.6) | 7.61 (3.56; 16.52) | <0.001 |
| others | 1972 (57.5) | |||||
| Patients with DRB1*04/x (DRB1*04 combining the DRB1*04:01/02/04/05 alleles, see text) | ||||||
| DRB1* | DQA1* | DQB1* | Controls | T1D Patients | Odds Ratio (CI) | p |
| n (%) | n (%) | |||||
| 01:01 | 01:01 | 05:01 | 291 (11.6) | 15 (25.0) | 2.55 (1.39; 4.53) | 0.001 |
| 01:02 | 01:01 | 05:01 | 36 (1.4) | 2 (3.3) | 2.37 (0.55; 9.34) | 0.221 |
| 04:01 | 03:02 | 03:01 | 44 (1.8) | 2 (3.3) | 1.94 (0.45; 7.44) | 0.291 |
| 07:01 | 02:01 | 02:02 | 260 (10.3) | 3 (5.0) | 0.46 (0.15; 1.35) | 0.276 |
| 07:01 | 02:01 | 03:03 | 100 (4.0) | 1 (1.7) | 0.41 (0.04; 2.24) | 0.730 |
| 08:01 | 04:01/02 | 04:02 | 202 (8.0) | 14 (23.3) | 3.48 (1.91; 6.38) | <0.001 |
| 08:04 | 04:01 | 04:02 | 8 (0.3) | 1 (1.7) | 5.31 (0.47; 35.27) | 0.192 |
| 11:01 | 05:05 | 03:01 | 254 (10.1) | 4 (6.7) | 0.64 (0.24; 1.69) | 0.515 |
| 11:03 | 05:05 | 03:01 | 28 (1.1) | 1 (1.7) | 1.51 (0.14; 9.12) | 0.497 |
| 12:01 | 05:05 | 03:01 | 53 (2.1) | 1 (1.7) | 0.79 (0.08; 4.46) | >0.999 |
| 13:01 | 01:03 | 06:03 | 209 (8.3) | 5 (8.3) | 1.00 (0.43; 2.39) | >0.999 |
| 13:02 | 01:02 | 06:04 | 118 (4.7) | 6 (10.0) | 2.26 (1.03; 5.30) | 0.058 |
| 13:03 | 05:05 | 03:01 | 40 (1.6) | 1 (1.7) | 1.05 (0.10; 6.08) | 0.623 |
| 16:01 | 01:02 | 05:02 | 90 (3.6) | 4 (6.7) | 1.92 (0.73; 5.00) | 0.278 |
| others | 781 (31.1) | |||||
The LMS method on page 3, l. 38 denotes the Lambda-Mu-Sigma method, not the least-mean-squares method. Alongside SAS, we used Graphpad Prism 7 with the Baptista Pike method for the estimation of confidence intervals (CI) for odds ratios.
In Table 1 for DRB1*14:01/54*-DQA1*01:01-DQB1*05:03, the OR is 0.00 (0; 0.39), p = 0.003, while for DRB1*15:01-DQA1*01:02-DQB1*06:02, the OR is 0.00 (0; 0.05), p < 0.001, and these were previously interchanged.
For clarification, we would like to add in brackets on page 6, ll. 7 f.: While 124 patients exhibited combinations of these, heterozygous (n = 104) or homozygous (n = 20, including two cases with a regular DR4 high-risk haplotype in combination with DRB1*04:05-DQA1*03:01-DQB1*02:02), 97 patients had one of these two high-risk haplotypes (‘moderate risk’) and only 12 cases (5.2%) did not show either one (‘low risk’).
The corrected Table 2 is given below. We accordingly reformatted the paragraph in the results section on page 6, ll. 13 ff.:
Of note, potentially protective alleles were missing, even in the second alleles. Genetic heterogeneity was low and most of the second heterozygous alleles were neutral and shared between the two groups. Additional susceptibility was conveyed by DRB1*01:01-DQA1*01:01-DQB1*05:01 and DRB1*08:01-DQA1*04:01/02-DQB1*04:02 in both groups, as well as DRB1*16:01-DQA1*01:02-DQB1*05:02 and less frequently DRB1*09:01-DQA1*03:02-DQB1*03:03 in the DRB1*03/x group, while DRB1*13:02-DQA1*01:02-DQB1*06:04 just failed significance in the DRB1*04/x group. Investigating the third group of 12 cases, which did not show either DR3-DQ2 or DR4-DQ8, the abovementioned five haplotypes were present in nine of 12 cases.
As significance is missing for the allele DRB1*13:02-DQA1*01:02-DQB1*06:04, it has to be deleted in the discussion section on page 10, l. 29.
Acknowledgments
We thank Jochem König, Institute for Medical Biostatisics, Epidemiology und Informatics, University Medicine of Johannes Gutenberg University Mainz (Germany) for pointing to inconsistencies in Table 2 leading to the correction of the estimates.