Mortality and Cancer in Pediatric-onset IBD
Mortality and Cancer in Pediatric-onset IBD
In this large population-based cohort of pediatric-onset IBD patients with a median follow-up of >11 years, global mortality risk was increased 1.4-fold but did not significantly differ from that of the general population, even after excluding the death because of car accident, yet there was a significant 3-fold increased risk of cancer.
The correlation between increased mortality rate and younger age at time of IBD diagnosis remains controversial. Two population-based studies reported diminished survival in patients diagnosed with CD later in life while a referral center-based study found no correlation between age at diagnosis and mortality ratios. Other population-based studies found that SMR was increased in patients diagnosed with CD as teenagers. A study from the United Kingdom demonstrated that patients aged 20–39 at study inception, and who were most likely diagnosed in adolescence, had a higher SMR. In a recent large population-based study from Denmark assessing trends in overall and cause-specific mortality among patients with IBD from 1982 to 2010, mortality was higher among patients diagnosed with either CD or UC at younger ages. Patients diagnosed with UC in childhood or adolescence had 2.15-fold (95% CI, 1.67–2.76) higher relative mortality than patients diagnosed with UC at age 60–79 years. For CD, mortality was 62% higher in patients diagnosed at age 0–19 years as compared with patients diagnosed at age 60–79 years. Overall IBD mortality was not significantly increased in our cohort but these data need to be interpreted with caution since our median follow-up was only a relatively short term of 11 years. In the Danish study, Kaplan–Meier plots for both diseases revealed that excess mortality in young individuals was primarily because of long-term increased risk of dying.
The cumulative probability of developing cancer was significantly increased in this young population with an overall 4% risk 17 years after initial IBD diagnosis. The variety of cancers found is consistent with the IBD literature. Two patients had a type of colon cancer that is particularly rare in young patients. Several studies have shown that the highest relative risk of colon cancer in IBD was observed in patients diagnosed younger than 25 years of age. Two patients, one with colon cancer and one with a cholangiocarcinoma had associated primary sclerosing cholangitis confirming the well-documented increased risk of neoplasia in this subgroup of patients. Interestingly, two patients who were both exposed to AZA had basal cell carcinoma at the ages of 17 and 26. In the CESAME cohort, ongoing and past exposure to thiopurines significantly increased the risk of non-melanoma skin cancer in patients with IBD, even before the age of 50.
An increased risk of intestinal carcinoid tumors has been reported in CD as well as an increased risk of cervical uterine dysplasia and cancer in women with IBD. This may be partly explained by the fact that women with IBD tend to have suboptimal Pap smear performance and increased rates of cervical dysplasia. Of note, the human papillomavirus status was unknown among children enrolled in the EPIMAD registry including those who developed cervical cancer and cancer of the prepuce. An important negative finding was the lack of lymphoma considering the increasing concern of lymphoproliferative disorders associated with the rising use of ISs and anti-TNF-α therapies in children and adolescents with IBD. Based on data from the French prospective observational cohort study called CESAME, the expected number of lymphoma cases calculated in our pediatric-onset IBD cohort was only 0.44 after a median 12-year follow-up (0.11 for UC and 0.36 for CD). Hence, it is not surprising that we did not observe any cases of lymphoma in this cohort of 698 IBD patients.
Two patients died from Niemann–Pick disease and Kinsbourne syndrome, both neurodegenerative disorders. The intriguing association between CD and Niemann–Pick disease has already been described and our data provide additional evidence of an association yet should be further explored.
The primary strength of this study was the assessment of a large pediatric-onset IBD population based study (n=698) covering a large area (9.3% of the French population) with well documented cases. The main limitation was the small number of recorded events (six deaths and nine cancers) so that we were not able to definitively evaluate the impact of medical therapies on the risk of cancer during the relatively short period of follow-up.
In conclusion, despite its overall reassuring message, this study should remind clinicians about the increased risk of cancer, which is especially worrisome in pediatric IBD, and the need for simple preventive measures such as endoscopic surveillance of the colon and protection against ultraviolet radiation with lifelong dermatologic screening in patients exposed to thiopurines and anti-TNFs.
Discussion
In this large population-based cohort of pediatric-onset IBD patients with a median follow-up of >11 years, global mortality risk was increased 1.4-fold but did not significantly differ from that of the general population, even after excluding the death because of car accident, yet there was a significant 3-fold increased risk of cancer.
The correlation between increased mortality rate and younger age at time of IBD diagnosis remains controversial. Two population-based studies reported diminished survival in patients diagnosed with CD later in life while a referral center-based study found no correlation between age at diagnosis and mortality ratios. Other population-based studies found that SMR was increased in patients diagnosed with CD as teenagers. A study from the United Kingdom demonstrated that patients aged 20–39 at study inception, and who were most likely diagnosed in adolescence, had a higher SMR. In a recent large population-based study from Denmark assessing trends in overall and cause-specific mortality among patients with IBD from 1982 to 2010, mortality was higher among patients diagnosed with either CD or UC at younger ages. Patients diagnosed with UC in childhood or adolescence had 2.15-fold (95% CI, 1.67–2.76) higher relative mortality than patients diagnosed with UC at age 60–79 years. For CD, mortality was 62% higher in patients diagnosed at age 0–19 years as compared with patients diagnosed at age 60–79 years. Overall IBD mortality was not significantly increased in our cohort but these data need to be interpreted with caution since our median follow-up was only a relatively short term of 11 years. In the Danish study, Kaplan–Meier plots for both diseases revealed that excess mortality in young individuals was primarily because of long-term increased risk of dying.
The cumulative probability of developing cancer was significantly increased in this young population with an overall 4% risk 17 years after initial IBD diagnosis. The variety of cancers found is consistent with the IBD literature. Two patients had a type of colon cancer that is particularly rare in young patients. Several studies have shown that the highest relative risk of colon cancer in IBD was observed in patients diagnosed younger than 25 years of age. Two patients, one with colon cancer and one with a cholangiocarcinoma had associated primary sclerosing cholangitis confirming the well-documented increased risk of neoplasia in this subgroup of patients. Interestingly, two patients who were both exposed to AZA had basal cell carcinoma at the ages of 17 and 26. In the CESAME cohort, ongoing and past exposure to thiopurines significantly increased the risk of non-melanoma skin cancer in patients with IBD, even before the age of 50.
An increased risk of intestinal carcinoid tumors has been reported in CD as well as an increased risk of cervical uterine dysplasia and cancer in women with IBD. This may be partly explained by the fact that women with IBD tend to have suboptimal Pap smear performance and increased rates of cervical dysplasia. Of note, the human papillomavirus status was unknown among children enrolled in the EPIMAD registry including those who developed cervical cancer and cancer of the prepuce. An important negative finding was the lack of lymphoma considering the increasing concern of lymphoproliferative disorders associated with the rising use of ISs and anti-TNF-α therapies in children and adolescents with IBD. Based on data from the French prospective observational cohort study called CESAME, the expected number of lymphoma cases calculated in our pediatric-onset IBD cohort was only 0.44 after a median 12-year follow-up (0.11 for UC and 0.36 for CD). Hence, it is not surprising that we did not observe any cases of lymphoma in this cohort of 698 IBD patients.
Two patients died from Niemann–Pick disease and Kinsbourne syndrome, both neurodegenerative disorders. The intriguing association between CD and Niemann–Pick disease has already been described and our data provide additional evidence of an association yet should be further explored.
The primary strength of this study was the assessment of a large pediatric-onset IBD population based study (n=698) covering a large area (9.3% of the French population) with well documented cases. The main limitation was the small number of recorded events (six deaths and nine cancers) so that we were not able to definitively evaluate the impact of medical therapies on the risk of cancer during the relatively short period of follow-up.
In conclusion, despite its overall reassuring message, this study should remind clinicians about the increased risk of cancer, which is especially worrisome in pediatric IBD, and the need for simple preventive measures such as endoscopic surveillance of the colon and protection against ultraviolet radiation with lifelong dermatologic screening in patients exposed to thiopurines and anti-TNFs.
Source...