Page 38 - New epidemiological and PSMA-expression based paradigms in salivary gland tumors

36
Chapter 2
Patient, tumor, and treatment characteristics: The bimodal age distribution in female SGPA incidence remains unexplained. Further research is needed to explore any hormone influences.
According to the literature, salivary gland tumors affect the parotid, submandibular, and minor glands in a ratio of 10:1:1 [1,33]. The ratio we found was 12:1:2, possibly because of an absence of selection bias in our data.
In our cohort, submandibular SGPAs were more common in women than in men, whereas minor salivary gland SGPAs were more common in men than in women. Since we found no previous mention of any sex differences in SGPA location, further research is needed to confirm and explain this finding.
As the PALGA database focuses on pathology, information on the type of surgery performed was often missing (42%). Recently, new insights about the benefits of standardized structured pathology reporting [34] have led to improved reporting practices for high-incidence cancers in Dutch laboratories. Hopefully, this systematic approach will be adopted for other diseases, too, including for SGPA.
Resection margins had not been recorded in 21% of cases. In a posthoc analysis, these cases turned out to have a 1.5-fold higher likelihood of recurrence, even after adjustment for gender, age, location and type of treatment. There may be several reasons why margin data are often missing. First, SGPAs are usually removed without complete margins of normal salivary gland tissue, for instance when they are close to the facial nerve. Second, covering (pseudo) capsules may be very thin, and multinodular growth patterns make it hard to determine whether any nodules have been left behind. Third, SGPAs are benign, so there is little priority in describing their margins, unless the pathology order holds a specific request to do so, along with sufficient clinical information.
Recurrence rates and malignant transformation: Whereas the 4.6% first-recurrence rate we found in patients with at least five years of follow-up (n=2 719) replicates previous findings [6], our 12% second-recurrence rate at five years is lower than the 14% stated in most papers (Additional Table D). However, some caution is needed here, as populations and follow-up periods vary between cohorts, and none of the figures have taken any clinical or mortality data into account.