If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Cancer Council Queensland, Spring Hill, Queensland, AustraliaMenzies Health Institute Queensland, Griffith University, Gold Coast, Queensland, AustraliaSchool of Mathematical Sciences, Queensland University of Technology, Queensland, Australia
Survivors of invasive melanoma have an increased risk of developing second primary cancers; however, similar risks associated with in situ melanoma have not been established. We evaluated 39,872 survivors of first primary in situ melanoma diagnosed from 1982 through 2012 in Queensland, Australia. Relative risk of second nonmelanoma primary cancers was estimated from standardized incidence ratios with 95% confidence intervals. A total of 4,823 (12%) in situ melanoma survivors developed a second primary cancer. A small increased risk (6%) compared with the general population was found. In those younger than 50 years, risk was increased by 14% for all cancers combined. In situ melanoma survivors had significantly increased risks of developing lip, thyroid, pancreatic, and brain cancers and decreased risks of head and neck, and lung cancers. Male in situ melanoma survivors had a significantly increased risk of prostate cancer; female survivors had an increased risk of thyroid cancer and lymphoid leukemia. Findings indicate that in situ melanoma may predict the diagnosis of certain second primary cancers. This altered risk may be due to biological, behavioral, or genetic factors or increased medical surveillance, and it requires further investigation, particularly among people younger than 50 years.
Over the past 30 years, studies have shown that the incidence of in situ melanoma is increasing in Western countries (Australia, Europe, and the United States) and at a faster rate than invasive melanomas (
), the evaluation of the burden and risk of developing a subsequent primary cancer among in situ melanoma survivors remains limited, because in situ melanoma is not routinely recorded or reported in many cancer registries. Regular follow-up and surveillance are recommended after a diagnosis of an invasive melanoma; however, because complete histological clearance of in situ melanoma is considered curative (
), little to no follow-up of this patient group is usually provided. The ability to better characterize the risk of a second primary cancer in survivors of in situ melanoma is critical for developing clinical surveillance plans to manage this increasingly large patient population (
). All of these studies have been conducted in the Northern Hemisphere, where there are lower incidence rates of melanoma and also lower sun intensity than are experienced in Australia; three of the studies report on US Surveillance, Epidemiology and End Results (SEER) population-based registry data (
). All studies found that in situ melanoma survivors had an increased risk of developing a second primary invasive melanoma, with some studies also showing in situ melanoma survivors to be at increased risk of certain second primary nonmelanoma cancers (
Here, we describe an analysis of the risk of second primary cancers in a large population-based sample of in situ melanoma survivors from Queensland, Australia. Unlike previous articles, this study gives insight into the extreme effects of UVR, because this population is predominantly composed of Caucasians living at a low latitude with the world’s highest incidence rate of melanoma (
In our study population of 39,872 in situ melanoma survivors, we had a total of 21,805 (55%) males and 18,067 females (Table 1). Almost a third (30%) were younger than 50 years at diagnosis of their first in situ melanoma and 24% were 70 years or older at diagnosis. Overall, the most common site was the upper and lower limbs (n = 15,912, 40%), followed by the trunk (n = 12,270, 31%) and head and neck (n = 11,124, 28%). Median follow-up time for all in situ melanoma survivors was 8.3 years (interquartile range = 4.6–13.9 years), varying from 6.3 years for those 70 years and older to 10.6 years for those younger than 50 years.
Table 1Characteristics of Patients with a First Primary In Situ Melanoma, Queensland, from 1982 through 2012
During this time, we identified 4,823 metachronous second primary invasive cancers (excluding invasive melanoma) in our study population, with a median time between first and second diagnosis of 5.7 years (interquartile range = 2.6–10.4 years). Compared against cancer incidence rates in the general population, the overall risk of a subsequent cancer among in situ melanoma survivors was marginally higher (standardized incidence rate [SIR] = 1.06, 95% confidence interval (CI) = 1.03–1.09) (Table 2). We found significantly increased risks overall for thyroid, lip, brain, and pancreatic cancers and lymphoid leukemia. In situ melanoma survivors were only half as likely to receive a diagnosis of head and neck cancer compared with the general population (SIR = 0.53, 95% confidence interval = 0.43–0.67), and also exhibited a 30% decreased relative risk for lung cancer.
Table 2Relative Risk of Selected Second Primary Invasive Cancers (Excluding Melanoma) after First Primary In Situ Melanoma by Sex in the Queensland Cancer Registry from 1982 through 2014
For males, a 35% increased relative risk was also found for prostate cancer (Table 2). Further analysis by sex showed that the increased relative risk for thyroid cancer applied to females with in situ melanoma but was not significant for males and that female survivors were also found to have a significantly increased relative risk of lymphoid leukemia, whereas this effect was not present for males.
We investigated the role of age at first primary in situ melanoma on risk of second invasive cancer (Table 3). The overall relative risk was 14% higher for those younger than 50 years and 5% higher for in situ melanoma survivors aged 50–69 or 70 years and older. For specific cancers increased relative risks were found for myeloma, thyroid cancer, lymphoid leukemia, and colorectal cancer for those younger than 50 years at first diagnosis that were not observed in the older age groups (apart from an increased relative risk of lymphoid leukemia in the 70 and over age group). In contrast, the significantly increased relative risks of pancreatic and lip cancers only applied among in situ melanoma survivors who were 70 years and older. Decreased relative risks were also found for head and neck cancers within each age group and for lung cancer among those aged either 50–69 years or 70 years and older.
Table 3Relative Risk of Second Primary Invasive Cancer (Excluding Melanoma) by Age at First Primary In Situ Melanoma in the Queensland Cancer Registry from 1982 through 2014
Results were reasonably consistent in terms of the effect of the anatomical site on the relative risk of second primary cancers (Table 4). Relative risks were significantly lower, irrespective of body site of the in situ melanoma for second primary head and neck cancers and lung cancer but were elevated across all body sites for prostate cancer among male survivors. In situ melanomas on the head and neck were associated with a significantly increased relative risk of lip cancer, whereas the elevated relative risk for lymphoid leukemia was most apparent after in situ melanomas on the upper and lower limbs.
Table 4Relative Risk of a Second Primary Invasive Cancer (Excluding Melanoma) by Anatomical Location of First Primary In Situ Melanoma in the Queensland Cancer Registry from 1982 through 2014
Finally, we compared second primary cancers for survivors of in situ melanoma (Figure 1) with survivors of invasive melanoma (Figure 2). A total of 57,227 patients with a first primary invasive melanoma were included, of whom 56% were males and for whom the median age of diagnosis was 56 years. Thickness was recorded for 88% of first primary invasive melanomas; most of these (71%) were thin melanomas (<1 mm), and 14% were 2 mm or thicker. Of the 7,166 invasive melanoma patients who were diagnosed with a metachronous second primary invasive cancer (excluding invasive melanoma), the median time between first and second diagnosis was 6.7 years (interquartile range = 3.0–12.5). The patterns were generally similar, although survivors of in situ melanoma had a significantly higher relative risk of having pancreatic cancer diagnosed, whereas the corresponding relative risk was neutral for invasive melanoma survivors. Conversely, significantly increased relative risks for colorectal and kidney cancers, non-Hodgkin lymphoma, and myeloma and a significantly lower relative risk for bladder cancer were found after invasive melanoma but not for in situ melanoma.
Discussion and Conclusion
To our knowledge, this is the first study investigating the risk of subsequent cancer development in survivors of in situ melanoma, conducted in a location in the Southern Hemisphere with high UVR/high melanoma incidence. In survivors of a first primary in situ melanoma, we found a significantly increased risk of a second primary lip, thyroid, brain, and prostate cancers and lymphoid leukemia and a significantly decreased risk of head and neck, and lung cancers. This reinforces previous studies showing that people with a diagnosis of in situ melanoma are not only at greater risk of subsequent invasive melanomas (
) found a 32% increased risk of second primary cancers in men with an in situ melanoma and a 35% increased risk of second primary cancers in women with an in situ melanoma, but this included second primary invasive melanoma. The SEER study also noted a decreased risk of second primary lung cancers and colorectal cancer in the in situ melanoma survivor cohort, similar to the results of our study, most likely because of their higher socioeconomic status and lower smoking prevalence. However, similar to the studies using SEER data, smoking exposure was not available through our registry data.
We found that age at diagnosis of first primary in situ melanoma was important to risk of subsequent cancer. In those younger than 50 years, risk was significantly increased by 14% for all cancers combined relative to their age-matched peers. Survivors who were younger than 50 years old at first diagnosis had a significantly increased relative risk of having prostate, colorectal, and thyroid cancers, as well as lymphoid leukemia and myeloma, whereas for those 70 years and older, the relative risks for colorectal and thyroid cancers and myeloma were either neutral or lower than in the age-matched general population. The mechanisms for this variation in age-related relative risk are not understood.
The reason for the altered pattern of risk of second cancers in survivors of primary in situ melanoma also remains unclear. It is possible that increased medical surveillance of these survivors may partially explain these outcomes; however, previous research (
) has shown significant variability in the skin examination follow-up of in situ melanoma patients, and this will affect the potential role of surveillance. The increased incidence of a second primary prostate cancer in survivors of in situ melanoma is consistent with this surveillance argument and is supported by a recent Australian study, which found a 25% increased risk of a second primary prostate cancer in men with localized cutaneous melanoma (
). We can assume that these survivors of in situ melanoma may have had significant lifetime exposure to sunlight, which beyond increasing the risk of melanoma, is known for other biological activity. Others have suggested that the altered pattern of subsequent primary cancer risk observed in invasive (as opposed to in situ) melanoma survivors may be related to clustering of lifestyle (sociocultural) risk factors in the same subgroups of the population (
), so the lower incidence of lung cancer and head and neck cancer in survivors of in situ melanoma may be somewhat expected. We could not evaluate lifestyle risk factors in our study population, so the extent to which these issues may have influenced our results remains unknown. Studies examining cancer risk among breast screening participants have found a lower risk of the so-called lifestyle cancers (lung, bladder, head, and neck) in screening participants, pointing toward a clustering of sociocultural factors affecting risk. Whether this is evidenced in melanoma survivors requires urgent further investigations.
Because we used population-based registry data, our ability to access individual information on in situ melanoma patients was limited, and we can only speculate on the causes of these altered risk patterns. The use of a population-based cancer registry design also comes with strengths; in particular, it provided long-term follow-up for second primary cancer occurrence, thus eliminating selection bias. Although it is not possible to determine the ethnicity of the in situ cohort, it is highly likely that the majority of the cohort is Caucasian. This may mean that extrapolation of our findings to other races and ethnicities should be approached with caution.
We provide, to our knowledge, the first evidence in a Southern Hemisphere, high-UVR location that in situ melanoma survivors have an increased relative risk of lip, pancreatic, thyroid, brain, and prostate cancers and lymphoid leukemia, and significantly decreased relative risk of head and neck, and lung cancers. These findings, in conjunction with similar findings from Northern Hemisphere cohorts, suggest that diagnosis with an in situ melanoma may be an important indicator of future cancer risk.
Follow-up care in Australia for those with diagnoses of in situ melanoma is unclear (
). Our findings showing a similarity in strength of associations with second primary cancers for those with in situ versus invasive melanoma (as shown in Figures 1 and 2) suggest that similar follow-up surveillance as that given for an invasive lesion should be offered to those with in situ melanoma diagnosis. Our findings have also identified that in situ melanoma survivors who are younger than 50 years are at increased risk of a second primary cancer compared with their peers in the general population. Further research is required to determine the mechanisms for altered risk in younger (<50 years) survivors and to understand the roles of medical surveillance and/or biological processes that lead to increased risk of certain subsequent cancers and decreased risk of other subsequent cancers among in situ melanoma survivors.
Materials and Methods
We used a retrospective cohort design for this study. Patient data were obtained through the population-based Queensland Cancer Registry. Eligible patients were those with a first primary in situ melanoma diagnosed between 1982 and 2012. Insitu melanoma was defined as cases in which the primary site was coded as skin (C44.0–C44.9) and histology as melanoma (8720–8799), according to the
. Only Queensland residents with a first primary in situ melanoma were included. We also excluded patients for whom the date of death was on or before the date of diagnosis (n = 63).
All diagnoses of a second primary invasive cancer (excluding invasive melanoma) were linked using unique dummy identifier numbers. We excluded invasive melanoma as the the evidence is clear regarding the increased melanoma risk after an in situ melanoma diagnosis (
). Synchronous cancers (diagnosed within 2 months of the in situ melanoma) were excluded to minimize detection bias. Third or more subsequent primary cancers were not considered. Follow-up was accumulated from 2 months after first diagnosis until December 31, 2014 (a potential minimum of 22 months and maximum of 33 years per patient), date of death, or date of diagnosis of a second primary cancer, whichever occurred first.
SIRs were calculated by dividing the observed number of cases of second primary cancer by the expected number of cases in the general population after matching by sex, age group, and calendar year. The SIR is used to estimate the risk of an in situ melanoma survivor developing a second primary malignancy relative to the incidence of cancer among the general population. We examined SIRs for all second cancers combined (excluding invasive melanoma) and individually for 16 common types of cancer. CIs for the SIRs were derived from the Poisson distribution and calculated at the 95% level of certainty, and P-values were obtained by the Rothman-Greenland method (