Idiopathic thrombocytopenic purpura, Helicobacter pylori infection, and HLA class II alleles

The interesting letter of Veneri et al gives us the opportunity for some remarks. First of all, the authors confirm the high prevalence of Helicobacter pylori infection in idiopathic thrombocytopenic purpura (ITP) patients. In a recent careful review of the literature, including an update of our case series, we found 112H pylori–infected patients out of a total of 193 ITP patients studied so far.1-1 Thus, by including the 24H pylori infected, out of the 39 ITP patients reported by Veneri et al, the prevalence of H pylori infection in ITP is 58.6%. Of course, this important association between ITP and a bacterium infection may prompt to investigate whether host genetic factors are involved in the pathogenesis of the disease and in susceptibility to infection. With respect to this, the study of human leukocyte antigens (HLAs) seems to be appropriate.

Veneri et al found that a low frequency of HLA-DRB1*11 and -DQB1*03 alleles characterizes the ITP patients compared with healthy controls. Moreover, a low frequency of such alleles seems to be typical ofH pylori–negative patients. This is a suggestive finding, but we would like to introduce a note of caution. The complexity of HLA system, the variability of H pylori strains, and the yet not well defined pathophysiology of ITP make this type of in vivo investigations very complicated.

In particular, in the literature there is a very large amount of reports dealing with possible correlations between diseases with underlying immune mechanisms and HLA system, with a great many alleles that correlate to certain features of the disease, while other alleles do not. For example, the HLA-DRB1*11 allele has been found at low frequency in hepatitis C virus (HCV)–positive Turkish patients,1-2 whereas that allele has been found at high frequency in French HCV-positive patients with mixed cryoglobulinemia and vasculitis.1-3 In HIV-positive patients the -DR*11 allele seems to represent a risk factor for the AIDS development and poor prognosis, but some reports have failed to confirm such an association between HLA and the disease.1-4 In aplastic anemia, HLA-DR2 has been found at increased frequency, particularly in patients with associated paroxysmal nocturnal hemoglobinuria.1-5 The findings about H pyloriinfection and HLA are contradictory. Some HLA alleles, like -DQB1*0602, have been found at high frequency in H pylori–positive patients favoring gastric cancer occurrence, whereas -DQB1*0301 seems to be protective against cancer in Taiwanese patients.1-6 In contrast, other alleles were found to be not associated with H pylori infection and cancer risk in Italian (-DQA1 and -DQB1) and German (-DQA1, -DQB1, and HLA class II) patients, respectively, whereas -DRB1*1501, -DQA1*01021, and -DQB1*0602 seem to be protective against H pylori infection in Japanese patients.1-7,1-8 Similarly, other different alleles were found at high or low frequency in Caucasian, Chinese, Spanish, Greek, and Japanese patients infected with H pylori.1-9-1-13 Similarly, as reported by Veneri et al, the few studies on ITP and H pylori infection revealed contradictory findings concerning a correlation with HLA alleles.

A better reassessment of these data and their possible clinical relevance should await for further rigorous studies on specific HLA alleles on very large series of ITP patients, which could necessarily take into account the racial differences of the population groups tested and also the H pylori–strain differences.