The presence of Panton-Valentine leukocidin (PVL) gene in methicillin-resistant Staphylococcus aureus (MRSA) infections has been associated with high severity and mortality rates. In this study we evaluated the effect of PVL on outcome in cancer patients with MRSA infections.We retrospectively reviewed the records of 173 cancer patients diagnosed with MRSA pneumonia (n = 47), skin and soft tissue infections (n = 77), and bacteremia (n = 49) between September 2003 and September 2007 at M. D. Anderson Cancer Center. We obtained demographic and clinical data and tested isolates for the presence of PVL. The data were compared between patients with PVL (+) and those with PVL (-) strains. Statistical methods for comparison included Cochran-Mantel-Haenszel tests, 2-way nonparametric analysis of variance, chi-square or Fisher exact tests, and Wilcoxon rank sum tests. All tests were 2-sided with a significance level of 0.05.Seventy patients with PVL (+) strains and 103 patients with PVL (-) strains were included in our study. Fewer PVL (+) patients had pneumonia than did PVL (-) patients (14% vs. 36%, p = 0.002). PVL (-) patients were more likely to have concomitant infections (35% vs. 17%, p = 0.012). The 2 groups were similar in terms of fever, sepsis, vasopressor use, mechanical ventilation, antibiotics response, infection relapse, death, and death due to MRSA. In a skin and soft tissue subset analysis, PVL (+) patients were more likely to have solid tumors (73% vs. 47%, p = 0.02) and less likely to have fever (20% vs. 44%, p = 0.02) and sepsis (12% vs. 36%, p = 0.013). There were no differences in outcome between patients with pneumonia and bacteremia; however, most patients with pneumonia were PVL (-) (79% vs. 21%). Among the 73 patients who received vancomycin and the 20 who received linezolid, there was no difference in response to treatment, regardless of PVL status or neutropenia. In conclusion, the presence of the PVL gene had no negative effect on cancer patients with health care-associated MRSA.