An association between Kaposi's sarcoma (KS) and malignant lymphoma ha
s been suspected for many years. Both cancers belong to the group of m
alignancies associated with immune suppression and have been known to
occur in the same individual. Accordingly, a common etiology has been
suspected. Through linkage within the Nordic cancer registries, we stu
died the occurrence of cancers in a population-based cohort of 741 pat
ients with classical KS. The relative risk of subsequent malignancies
was expressed as the ratio of the observed numbers of cancer to the nu
mbers expected based on age-, sex-, period- and country-specific incid
ence rates, i.e., the standardized incidence ratio (SIR). A total of 1
04 cancers was observed during 5,802 person-years of follow-up, which
was close to the expected 98.8 cases (SIR, 1.05). During the first yea
r of follow-up, 3 lymphomas were observed, which is in significant exc
ess of the 0.2 lymphomas expected (SIR, 13.0). In contrast, no lymphom
as occurred in the period beyond the first year of follow-up vs. 2.3 e
xpected. Cancers of the buccal cavity and pharynx (SIR, 10.6; n = 4) a
nd of the colon (SIR, 2.7; n = 7) occurred in excess among women but n
ot among men. Accordingly, our results indicate that patients with cla
ssical KS are not at increased risk of cancer in general. In particula
r, the overall risk of lymphomas was not significantly increased. The
high relative risk of malignant lymphoma immediately after KS was base
d on a limited number of cases, and this observation is unlikely to in
dicate a common etiology. (C) 1997 Wiley-Liss, Inc.