It was with great interest that we read the manuscript concerning the role of bacille Calmette-Guérin (BCG) heat-shock protein 65 (HSP-65) in the high-grade bladder cancer patient outcome prediction, by Ardelt et al, recently published in Cancer.1 In this interesting and well-done study, the authors gave special attention to a fundamental aspect of high-grade bladder cancer patient management: the prediction of response to BCG therapy. Several researchers have argued that some molecular markers should be used for predicting the clinical outcome of patients affected by nonmuscle invasive bladder cancer (NMIBC) and treated with intravesical BCG, but as yet no marker has been identified in everyday clinical practice,2 as highlighted also by Ardelt et al.1 They found that increasing IgA and IgG anti-HSP-65 titers specifically predicted a positive patient outcome in a cohort of patients with bladder cancer. Ardelt et al also showed the possible role of antibody production against Mycobacteriumbovis BCG HSP-65 as a serologic marker for the predictive outcome of BCG immunotherapy.1
These data are extremely fascinating. In our previous study, we investigated the role of loss of heterozygosity (LOH) on interferon-alpha (IFN-α) locus for predicting the response to BCG therapy and were able to show a significant correlation between the presence of LOH on chromosome 18 and status at follow-up.3 In fact, we know that the mechanism of BCG is mediated by TH1 cytokines, such as IL-12 and IFN-alfa, which activate natural killer (NK) cells.2 Moreover, a correlation between HSPs and T-cell function has been hypothesized and supported by the finding that HSPs are expressed on the cell surface, play an important role in antibody assembly and antigen presentation, and are recognized by T-cells including NK cells.4
We pose 3 questions. 1) Can we suppose that the increased IgA and IgG anti-HSP-65 titers are linked to a normal expression of INF-alfa locus and, perhaps, to a normal INF-alfa function? Moreover, can we hypothesize the presence of an additional humoral antitumour response? If yes, can HSPs be considered the link between these 2 mechanisms?