Despite the high prevalence and impact on quality of life, costs, and survival, there are still unresolved issues regarding diabetic polyneuropathy (DPN): the lack of definite knowledge of its pathogenesis; the limited preventive action of glycaemic control in type 2 diabetes; and the unavailability of evidence-based effective disease-modifying treatment. How can genetics provide the tools to address these gaps? Ziegler et al for the GDS Group explore the novel hypothesis that genetic variability in transketolase (TKT) might contribute to susceptibility to DPN in patients with newly diagnosed type 1 and type 2 diabetes (well characterised for DPN). Transketolase diverts excess glycolytic metabolites from the hexosamine, protein kinase C, and advanced glycation endproduct pathways to the pentose phosphate pathway, with a protective effect against hyperglycaemia-induced damage. Moreover, thiamine and its derivative benfotiamine are among the few disease-modifying agents still under consideration as DPN treatment. The authors find significant associations of single-nucleotide polymorphisms of the TKT gene with the Total Symptom Score and thermal thresholds, in particular in male participants with type 2 diabetes. Moreover, they measure plasma methylglyoxal (a glycating agent, whose availability is hindered by TKT) without however finding a relation with TKT single-nucleotide polymorphisms. The link found between TKT genetic variability and nerve function measures is considered here in the context of DPN genetic studies and of experimental and clinical findings regarding thiamine and benfotiamine. The conclusion is that available data supports the decision to maintain focus on both the search for DPN genetic biomarkers and the therapeutic attempts to target thiamine, TKT, and methylglyoxal.