Questionable evidence of increasing incidence of invasive penile cancer in Denmark: authors reply

Abstract

We appreciate the interest and comments of Dr. Frisch et al. on our paper about the incidence of penile cancer and high-grade penile intraepithelial neoplasia grades 2 and 3 (PIN 2/3) in Denmark. As they correctly pointed out, we conducted the study using data from the Danish Cancer Registry as well as from the Danish Pathology Data Bank (PDB). Even though we have previously published incidence studies including only data from the Danish Cancer Registry, the addition of data from the PDB is based on our recent observation that at least for some cancers this will add a number of incident cases not identified in the Danish Cancer Registry. As penile cancer is a rare cancer, we found it important to take advantage of the existence of multiple data sources. With this rare cancer, the analyses may be especially vulnerable to underreporting and by including several data sources, we tried to reduce this potential bias. In stead of only displaying the combined results, we chose to present the data separately to be as transparent as possible. We fully agree with Dr. Frisch et al. that the PDB also has underreporting in the initial period and is most likely not complete until around the mid-1990s, and we also agree that there may be too much focus on the trend going all the way back from 1978 in our paper and that the main interpretation of the incidence of penile cancer should focus on the more recent years in the analyses in our paper. As mentioned by Dr. Frisch et al., a Dutch study has also looked at incidence of penile carcinoma and carcinoma in situ [1]. We have performed additional analyses so that comparison between the two studies is straighter forward. In line with that study, we looked at all penile carcinomas, all invasive penile carcinomas and all non-invasive (in situ) penile carcinomas, and we found that the incidence of all penile carcinomas increased from 1.2 in 1997–1998 to 1.8 in 2007–2008. However, the increase was mostly pronounced for the noninvasive penile carcinomas with an estimated average annual percentage change of 1.2 % (p = 0.03), whereas the estimated annual percentage change for invasive carcinomas in this period (0.9 %) did not reach statistical significance (p = 0.34) pointing to an increase in incidence especially of carcinoma in situ. These results are in line with the results of the Dutch study [1]. In our study, we calculated the incidence of the PIN 2/3 lesions as lesions with [2-year interval and found that the incidence increased from 0.5 to 0.9 per 100,000 men-years in the study period. A more conservative way to assess the incidence of these pre-invasive lesions is, of course, to count only the first episode of PIN 2/3 in each man. When we performed this analysis, we found virtually the same result with the incidence increasing from 0.47 to 0.82 per 100,000 men-years (Fig. 1). We again thank Dr. Frisch et al. for bringing some of their insightful thoughts to our paper.