Men with no sons have an increased risk of prostate cancer in relation to those with at least one son:

The researchers in the Mailman School's Department of Epidemiology analyzed the relative risk of prostate cancer by the sex of offspring among fathers registered in a family-based research cohort in Israel. From this cohort of 38,934 men, followed from the birth of their offspring (in 1964 through 1976) until 2005, the authors conclude that genes on the Y chromosome may be involved in prostate cancer risk in this population.

"We surveyed vital status and cancer incidence, and found a strong trend for a decrease in prostate cancer risk as the number of sons increased," said Susan Harlap, MD, professor of clinical Epidemiology at the Mailman School of Public Health, and the leader of the research team. "We anticipate that this finding will have a significant impact on the direction of research in this field going forward." Overall, there was a 40% increase in prostate cancer in men lacking sons.

...

The researchers also looked at men lacking daughters. For example, in men with exactly two offspring, those with no daughters had an 11% increase in the incidence of prostate cancer, and those with no sons had a 47% increase, compared with men who had one son and one daughter. "The increased risk of prostate cancer in men with no daughters is probably due to chance," says Dr. Harlap, "but it might indicate a problem with a gene on the X-chromosome."

The researchers speculate that failure to have sons might be indicative of some problem with the man's Y chromosome that is cosegregating with prostate cancer risk.

The finding that men with no daughters also have increased risk is interesting, but the effect size is so small I wonder if it is significant.