Response to Dr. Friaza submission: “Prevalence of dihydropteroate synthase mutations in Spanish patients with HIV-associated Pneumocystis pneumonia”

Abstract

We thank Dr. Friaza and colleagues for their letter. We share their interest in gaining a better understanding of the prevalence of dihydropteroate synthase (DHPS) mutations in Spanish patients with Pneumocystis jirovecii pneumonia (PcP). The reported prevalences of DHPS mutations, possibly associated with drug resistance, vary widely geographically (Kazanjian et al., 1998, 2000, 2004; Beard et al., 2000; Huang et al., 2000). Knowing the prevalence of these mutations in Spanish PCP patients could have important public health implications. As pointed out by Dr. Friaza, we incorrectly cited their reported prevalence of Pneumocystis DHPS mutations (Alvarez-Martinez et al., 2008). Nevertheless, their reported prevalence of DHPS mutations was 35.5% overall and 45.5% in HIV-infected patients (Montes-Cano et al., 2004). This prevalence is much higher than the 3.7% reported in our study (Alvarez-Martinez et al., 2008). Dr. Friaza and colleagues suggested that this difference was due to analytical methods. However, both studies used purified PCR amplicon, not subcloned DNA, so it is unlikely that minority variants (either wild type or mutant) were overrepresented in either study (Alvarez et al., 2006; Montes-Cano et al., 2004). Previous studies have not found disparities between the RFLP method used in their study and direct DNA sequence analysis (used in our study) when conducted on the same samples. We suspect that the disparity is due to sampling bias. The study by Montes-Cano et al. (2004) included only 19 HIVpositive patients at a single health facility. Our study was conducted on 188 HIV-positive patients in 12 sites throughout Spain. Thus, we suspect that our results may be more representative of the overall Spanish HIV patient population. The results of both papers can be harmonized most easily considering another recent observation. We recently showed that there was a very high prevalence of DHPS mutations in PcP episodes in Spanish HIV patients diagnosed before the introduction of cART. The prevalence of DHPS mutations in 163 PcP episodes in HIV-1–infected patients diagnosed between 1989 and 1995 was 33% compared to only 6% in the cART period between 2001 and 2004 (Alvarez et al., 2005). This difference was probably