Recognizing Complications Dig Dis 2014;32:455–462 DOI: 10.1159/000358152
Inflammatory Bowel Disease and Neoplasia in Children Paul A. Rufo Center for Inflammatory Bowel Disease, Boston Children’s Hospital, Boston, Mass., USA
Abstract Advances in our understanding of the pathophysiology of ulcerative colitis and Crohn’s disease have led to the widespread use of increasingly potent immunosuppressive therapies. This has greatly benefited the majority of patients with inflammatory bowel disease (IBD) and has resulted in improved overall clinical outcomes and quality of life. However, a growing body of data now indicates that long-term use of these agents may at the same time place patients at risk for a number of adverse effects, including colorectal and skin cancer, as well as lymphoma. Children and adolescents may be at particular cumulative risk for the development of these malignancies as a result of their young age at diagnosis, often increased disease extent and severity, and greater lifetime exposure to immunosuppressive agents. More recent epidemiologic studies are now identifying specific genetic and treatment-related factors that may place patients at increased risk for the development of IBD-related cancer. Improved understanding of these risk factors should contribute to a more rational approach to the pharmacologic man-
© 2014 S. Karger AG, Basel 0257–2753/14/0324–0455$39.50/0 E-Mail [email protected] www.karger.com/ddi
agement of children and young adults with IBD. Similarly, clinicians will be better able to counsel patients about the risks and benefits associated with specific therapies and develop improved monitoring guidelines to reduce the incidence of these rare but often severe oncologic complications. © 2014 S. Karger AG, Basel
Colorectal Cancer in Children with Ulcerative Colitis
The increased cumulative risk of colorectal cancer (CRC) observed in patients with inflammatory bowel disease (IBD), and particularly those with ulcerative colitis (UC), has been the subject of considerable discussion and research. The pathogenesis of CRC in patients with IBD has not yet been fully elucidated, and this has precluded the development of mechanistic evidence-based screening and prevention programs. Existing data suggest that a combination of host factors, including increased cell turnover in chronic intestinal inflammatory states, as well as the possibility of an underlying increased risk of malignancy in patients with IBD, may be playing a role in the development of neoplasia in these patients. Epidemiologic studies have further underscored the relevance of the Paul A. Rufo, MD, MMSc Center for Inflammatory Bowel Disease Boston Children’s Hospital, 300 Longwood Avenue, HU 103.3 Boston, MA 02115 (USA) E-Mail paul.rufo @ childrens.harvard.edu
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
Key Words Colorectal cancer · Immunosuppression · Lymphoma · Pediatric neoplasia · Skin cancer
456
Dig Dis 2014;32:455–462 DOI: 10.1159/000358152
studies. Variance in study outcome may also be explained by the timing and duration during which studies were completed. Improved surveillance strategies resulting in an opportunity for earlier and more effective surgical intervention, as well as the more widespread application of immunomodulatory and biologic therapies in the past 2 decades, has represented a paradigm shift from previous clinical practice and calls into question any head-to-head comparison of temporally conflicted studies discussing long-term morbidity and mortality outcomes. Finally, differing study endpoints have complicated the analysis of CRC risk in patients with IBD. While earlier case report studies focused on the incidence of malignancy and longterm survival in patients with CRC, more recent studies often use the development of mucosal dysplasia as the primary outcome variable. Further complicating this picture is the often frequent difficulty in defining the criteria for low- and high-grade dysplasia across centers in the same study and across different research studies. Goel and Shanks [2] reported on a cohort of 25 children and young adults that had been diagnosed with UC over a 40-year period extending from 1931 to 1972. The average follow-up was 24 years. They reported only 1 patient (4%) with CRC in this series. A review of the literature at the time revealed that a similar rate had been reported in a number of small- to medium-sized case series between 1950 and 1973. In studies that included from 25 to 137 children, an age at diagnosis that ranged from 5 to 16 years, a duration of UC from 5 to 25 years, and an average follow-up that ranged from 2.5 to 22 years, the incidence of carcinoma ranged from 3.2 to 4.6%. Perhaps the largest series of its type was completed by Devroede et al. [3] in their review of the records of 396 patients, all aged 14 or less, that were diagnosed with UC at the Mayo clinic between 1919 and 1965. Comprehensive follow-up was completed through a combination of phone interviews, letters, or direct contact with the primary-care providers of these patients. Their data indicated that CRC risk was very low for the first 10 years after diagnosis, increased steadily thereafter, and reached a cumulative probability of 43% by 35 years after diagnosis. Their data demonstrated no gender gap in CRC incidence. However, age at diagnosis was a strong predictor, and children diagnosed between 5 and 9 years of age were found to be at much greater risk for the development of CRC than those diagnosed later (p = 0.02). Extent of involvement emerged as a relevant risk factor as well, and those with pancolitis proved to be at significantly greater risk for the development of CRC than those with either left-sided disease or in the rectum only (p = 0.01). Rufo
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
duration, severity, and extent of colonic inflammation as risk factors for CRC. In addition, environmental factors including long-term use of immunosuppressive agents, as well as environmental factors including changes in the host microbiome, are likely contributing to the development of CRC pathogenesis. As such, clinicians have been left to rely on dysplasia surveillance and early cancer detection as the primary treatment strategies in the approach to their patient at risk for CRC. No randomized study has been completed to assess the impact of scheduled surveillance on the long-term mortality and morbidity of CRC, and it is unlikely that ethic standards will permit the execution of such a study in the future. Nonetheless, there is a general consensus that proactive screening programs enable clinicians to identify CRC lesions at earlier stages, and at a time when curative resection is still possible. CRC is a rare complication of IBD in childhood. However, it is important for clinicians caring for children and young adults with IBD to be aware of this complication, discuss the relative risks to their patients and parents, and to make sure that appropriate CRC screening programs are initiated in their patients at especially high risk. Studies addressing the incidence of CRC have frequently resulted in controversy and conflicting findings [1]. Much of the disparity in outcome has likely been related to systematic biases originating in competing study designs and populations. Conclusions drawn from databases from tertiary-care centers are likely to disproportionately reflect the outcome of patients with particularly complicated histories, including those exposed to longterm use of systemic immunosuppressive therapies. In contrast, findings drawn from large population-based studies may be biased by specific genetic variables as well as ethnic and cultural variables, including diet, levels of physical activity, and the prevalence of smoking. The methodology of the type of study performed can similarly be a source of result diversity. Case series are typically completed with well-defined populations, often with direct follow-up over several years or decades. However, these studies are subject to bias arising from specific center practices concerning screening and pharmacologic therapies. In addition, the relatively small number of subjects in these studies may result in insufficient power to reach definitive conclusions. Meta-analyses, in contrast, can compensate for this weakness by pooling the results from a number of related studies. As such, outcomes from meta-analyses are going to rest heavily upon the investigator’s a priori decisions concerning inclusion criteria, diagnostic approach, length and quality of follow-up, as well as the comparability of outcome variables across
Civil Registration System, Jess et al. [6] reported a fall in the RR of CRC from 1.3 in 1979–1988 to 0.57 in 1999– 2008. However, populations of patients diagnosed at younger age and those with more extensive disease or concomitant primary sclerosing cholangitis (PSC) remained at increased risk. Ongoing epidemiologic studies will continue to expand our understanding of the interplay between age, disease extent, and perhaps response to therapy as independent risk factors for the development of CRC. At the same time, ongoing genomic and epigenetic studies are underway to elucidate the mechanism(s) driving the increased risk of CRC observed in our patients with UC. The confluence of these efforts will ultimately provide clinicians with the information necessary to identify individual patients at the greatest risk for the development of CRC, and contribute to the development of more effective invasive and noninvasive markers of CRC and more effective chemoprophylactic strategies. Until this level of personalized medicine is within reach, clinicians must decide how and when to include a discussion of CRC when counseling individual patients with UC and their parents, and how to manage CRC risk in their clinical practice. Clinical experience, in conjunction with data from >40 years of aggregate research, demonstrates that the incidence of CRC in children and young adults is quite small. However, the severe nature of this complication mandates its inclusion in any discussions concerning longterm risks associated with UC. Though our present understanding of CRC is incomplete, available data do afford clinicians with the opportunity to begin to consider individual patients along particular demographic and clinical dimensions in efforts to identify those at particular risk for the development of CRC. • How old was the patient when UC was diagnosed? Patients that were diagnosed with UC as children or young adults (age 0–19 years) may have an RR of developing CRC that reaches 43.85. In contrast, the RR falls to 2.65 if patients were diagnosed with UC between ages of 20–39 years and to background if they were diagnosed with UC after the age of 60 years. • How long has the patient carried a diagnosis of UC? The risk of developing CRC appears to be related to the length of time that a patient has been experiencing chronic intestinal inflammation. As such, surveillance strategies should be based on the interval since their diagnosis and not their present chronologic age [1]. • How extensive is the patient’s disease? Multiple studies have demonstrated that the risk of CRC is highest in those patients with pancolitis. Patients with disease
IBD and Neoplasia in Children
Dig Dis 2014;32:455–462 DOI: 10.1159/000358152
457
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
Eaden et al. [4] completed a meta-analysis in an effort to assess the risk of CRC in patients with ulcerative colitis. They were able to identify 26 studies from the adult literature and an additional 5 studies that estimated both the incidence of CRC in children (average age at diagnosis was 10 years) with UC and reported duration of follow-up (average 12 years). They reported a cumulative incidence of CRC for adults diagnosed with UC at 4.4, 8.6, and 12.7% at 10, 20, and 30 years after diagnosis, respectively. Data from the five eligible studies that included children reported similar, albeit higher (5.5, 10.8, and 15.7%) cumulative risk rates for CRC. Interestingly, data from this study also suggested a geographic influence with greater rates of CRC observed in studies from the UK and United States, compared to those from Scandinavia. As noted above, it is unclear if these differences are the results of underlying genetic or dietary covariance, environmental exposures, or medical practice styles. Ekbom et al. [5] completed a population-based study in Sweden. They reviewed the medical charts of 613 patients (97 aged 50 years, respectively). The SIR for CRC also dropped steadily in cohorts diagnosed with UC with increasing age: 60 years (0.12). There is some suggestion that the incidence of CRC in patients with UC may be decreasing over time. It remains speculative as to whether or not this apparent decrease is related to improved surveillance strategies, more aggressive pharmacologic management, or earlier and more effective surgical approaches to the management of patients with UC. These more recent findings support the hypothesis that the translation of any conclusions drawn from any single study data must also take into account the interval being studied. In an especially large and comprehensive population-based study employing the Danish
Skin Cancer in Children with IBD
Skin cancers present a growing health burden in the US and abroad. In general, these cancers are separated into melanoma and nonmelanoma conditions. Melanomas arise from melanin-producing pigment cells found in the basal layer of the epidermis. They can occur in both sun-exposed and nonexposed parts of the body. Melanomas can be very aggressive. They are invasive and can metastasize early and before there is an opportunity for curative resection. Nonmelanoma skin cancers (NMSC) 458
Dig Dis 2014;32:455–462 DOI: 10.1159/000358152
include basal (BCC) and squamous cell carcinomas (SCC) [10]. Both NMSC forms typically arise in sun-exposed areas. BCC are much more common, represent 74% of all new diagnoses, and arise from cells in the lower part of the epidermis. BCC tend to be more superficial, indolent, and amenable to surgical resection. In contrast, SCC are derived from superficial keratinocytes, have a greater tendency to be invasive, and are more likely to metastasize early and recur [11]. Current data indicate that skin cancers are on the rise, and despite intensive marketing of sun-blocking products, the incidence of all forms of skin cancer have increased by 1.9% per year, among both men and woman, in the interval between 2000 and 2009. The mortality related to skin cancer has similarly increased by approximately 1% per year [12]. While gender has a modest impact on the risk of developing skin cancer, there appears to be a much more significant racial impact. While skin cancer rates have climbed in Caucasian populations, incidence and mortality rates have remained relatively stable over the same interval for darker-skinned (black and Hispanic) individuals. Sun exposure drives the development of skin cancer and, as such, geography plays a significant role in the development of these malignancies. With respect to NMSC, the risk appears to be related to the chronicity or cumulative effect of sun exposure [13]. In contrast, the risk of melanoma appears to be much more related to the intensity of intermittent exposures, such as that resulting in blistering sunburns [14]. Sun exposure generally varies with latitude and has been standardized by the World Health Organization in the form of an international UV index. UV indices vary from about 3.5 in Paris and Vancouver and rise to 11.5 in more southern latitudes, including Bangkok (10.5) and Singapore (11.5) [15]. The concerns about the possibility of an increased risk of skin cancer in patients with IBD, especially those receiving immunosuppressive agents, arose from earlier reports in the transplant literature. Solid organ transplant recipients, especially those receiving the same immunomodulatory agents now frequently used in the long-term management of both Crohn’s disease (CD) and UC, appear to be especially prone to the development of NMSC, and the malignancies in these patients tend to occur earlier and are more aggressive [16]. The risk of SCC is especially high and may be up to 65 times greater than in the general population [17]. The mechanism(s) underlying the increased risk of skin cancer in patients receiving immunosuppressive therapy remain incompletely understood. Existing in vitro data demonstrate that thiopurine Rufo
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
primarily on the left side of their colon are at intermediate risk. In most studies, patients with limited proctitis appear to be at no increased risk for the development of CRC than age-matched peers [1]. It remains unclear if the presence of backwash ileitis is an independent risk factor for the development of CRC [7]. • Does the patient have PSC? Population-based studies indicated that patients with PSC and IBD are at considerably greater risk (RR: 9.13) compared to patients with UC and no PSC [6]. A meta-analysis similarly supported this correlation and reported an odds ratio (OR) of 4.79 [8]. Data further suggest that patients with PSC are at particular risk for the early development of CRC [9]. For unclear reasons, CRC in patients with PSC is more likely to occur on the right side of the colon. In contrast, right-sided CRC is noted in only 38% of patients with IBD and no PSC. As such, current guidelines indicate that annual full surveillance colonoscopic evaluations should be performed in all patients with UC and PSC within 2 years of diagnosis. • Does the patient have a family history of CRC? Existing data suggest that a genetic context can serve as an additional risk modifier for the development of CRC in patients with UC. If a patient with UC has a first-degree relative with CRC, their RR of CRC doubles. The risk increases 9-fold if the family member of the UC patient was
Inflammatory Bowel Disease and Neoplasia in Children Paul A. Rufo Center for Inflammatory Bowel Disease, Boston Children’s Hospital, Boston, Mass., USA
Abstract Advances in our understanding of the pathophysiology of ulcerative colitis and Crohn’s disease have led to the widespread use of increasingly potent immunosuppressive therapies. This has greatly benefited the majority of patients with inflammatory bowel disease (IBD) and has resulted in improved overall clinical outcomes and quality of life. However, a growing body of data now indicates that long-term use of these agents may at the same time place patients at risk for a number of adverse effects, including colorectal and skin cancer, as well as lymphoma. Children and adolescents may be at particular cumulative risk for the development of these malignancies as a result of their young age at diagnosis, often increased disease extent and severity, and greater lifetime exposure to immunosuppressive agents. More recent epidemiologic studies are now identifying specific genetic and treatment-related factors that may place patients at increased risk for the development of IBD-related cancer. Improved understanding of these risk factors should contribute to a more rational approach to the pharmacologic man-
© 2014 S. Karger AG, Basel 0257–2753/14/0324–0455$39.50/0 E-Mail [email protected] www.karger.com/ddi
agement of children and young adults with IBD. Similarly, clinicians will be better able to counsel patients about the risks and benefits associated with specific therapies and develop improved monitoring guidelines to reduce the incidence of these rare but often severe oncologic complications. © 2014 S. Karger AG, Basel
Colorectal Cancer in Children with Ulcerative Colitis
The increased cumulative risk of colorectal cancer (CRC) observed in patients with inflammatory bowel disease (IBD), and particularly those with ulcerative colitis (UC), has been the subject of considerable discussion and research. The pathogenesis of CRC in patients with IBD has not yet been fully elucidated, and this has precluded the development of mechanistic evidence-based screening and prevention programs. Existing data suggest that a combination of host factors, including increased cell turnover in chronic intestinal inflammatory states, as well as the possibility of an underlying increased risk of malignancy in patients with IBD, may be playing a role in the development of neoplasia in these patients. Epidemiologic studies have further underscored the relevance of the Paul A. Rufo, MD, MMSc Center for Inflammatory Bowel Disease Boston Children’s Hospital, 300 Longwood Avenue, HU 103.3 Boston, MA 02115 (USA) E-Mail paul.rufo @ childrens.harvard.edu
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
Key Words Colorectal cancer · Immunosuppression · Lymphoma · Pediatric neoplasia · Skin cancer
456
Dig Dis 2014;32:455–462 DOI: 10.1159/000358152
studies. Variance in study outcome may also be explained by the timing and duration during which studies were completed. Improved surveillance strategies resulting in an opportunity for earlier and more effective surgical intervention, as well as the more widespread application of immunomodulatory and biologic therapies in the past 2 decades, has represented a paradigm shift from previous clinical practice and calls into question any head-to-head comparison of temporally conflicted studies discussing long-term morbidity and mortality outcomes. Finally, differing study endpoints have complicated the analysis of CRC risk in patients with IBD. While earlier case report studies focused on the incidence of malignancy and longterm survival in patients with CRC, more recent studies often use the development of mucosal dysplasia as the primary outcome variable. Further complicating this picture is the often frequent difficulty in defining the criteria for low- and high-grade dysplasia across centers in the same study and across different research studies. Goel and Shanks [2] reported on a cohort of 25 children and young adults that had been diagnosed with UC over a 40-year period extending from 1931 to 1972. The average follow-up was 24 years. They reported only 1 patient (4%) with CRC in this series. A review of the literature at the time revealed that a similar rate had been reported in a number of small- to medium-sized case series between 1950 and 1973. In studies that included from 25 to 137 children, an age at diagnosis that ranged from 5 to 16 years, a duration of UC from 5 to 25 years, and an average follow-up that ranged from 2.5 to 22 years, the incidence of carcinoma ranged from 3.2 to 4.6%. Perhaps the largest series of its type was completed by Devroede et al. [3] in their review of the records of 396 patients, all aged 14 or less, that were diagnosed with UC at the Mayo clinic between 1919 and 1965. Comprehensive follow-up was completed through a combination of phone interviews, letters, or direct contact with the primary-care providers of these patients. Their data indicated that CRC risk was very low for the first 10 years after diagnosis, increased steadily thereafter, and reached a cumulative probability of 43% by 35 years after diagnosis. Their data demonstrated no gender gap in CRC incidence. However, age at diagnosis was a strong predictor, and children diagnosed between 5 and 9 years of age were found to be at much greater risk for the development of CRC than those diagnosed later (p = 0.02). Extent of involvement emerged as a relevant risk factor as well, and those with pancolitis proved to be at significantly greater risk for the development of CRC than those with either left-sided disease or in the rectum only (p = 0.01). Rufo
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
duration, severity, and extent of colonic inflammation as risk factors for CRC. In addition, environmental factors including long-term use of immunosuppressive agents, as well as environmental factors including changes in the host microbiome, are likely contributing to the development of CRC pathogenesis. As such, clinicians have been left to rely on dysplasia surveillance and early cancer detection as the primary treatment strategies in the approach to their patient at risk for CRC. No randomized study has been completed to assess the impact of scheduled surveillance on the long-term mortality and morbidity of CRC, and it is unlikely that ethic standards will permit the execution of such a study in the future. Nonetheless, there is a general consensus that proactive screening programs enable clinicians to identify CRC lesions at earlier stages, and at a time when curative resection is still possible. CRC is a rare complication of IBD in childhood. However, it is important for clinicians caring for children and young adults with IBD to be aware of this complication, discuss the relative risks to their patients and parents, and to make sure that appropriate CRC screening programs are initiated in their patients at especially high risk. Studies addressing the incidence of CRC have frequently resulted in controversy and conflicting findings [1]. Much of the disparity in outcome has likely been related to systematic biases originating in competing study designs and populations. Conclusions drawn from databases from tertiary-care centers are likely to disproportionately reflect the outcome of patients with particularly complicated histories, including those exposed to longterm use of systemic immunosuppressive therapies. In contrast, findings drawn from large population-based studies may be biased by specific genetic variables as well as ethnic and cultural variables, including diet, levels of physical activity, and the prevalence of smoking. The methodology of the type of study performed can similarly be a source of result diversity. Case series are typically completed with well-defined populations, often with direct follow-up over several years or decades. However, these studies are subject to bias arising from specific center practices concerning screening and pharmacologic therapies. In addition, the relatively small number of subjects in these studies may result in insufficient power to reach definitive conclusions. Meta-analyses, in contrast, can compensate for this weakness by pooling the results from a number of related studies. As such, outcomes from meta-analyses are going to rest heavily upon the investigator’s a priori decisions concerning inclusion criteria, diagnostic approach, length and quality of follow-up, as well as the comparability of outcome variables across
Civil Registration System, Jess et al. [6] reported a fall in the RR of CRC from 1.3 in 1979–1988 to 0.57 in 1999– 2008. However, populations of patients diagnosed at younger age and those with more extensive disease or concomitant primary sclerosing cholangitis (PSC) remained at increased risk. Ongoing epidemiologic studies will continue to expand our understanding of the interplay between age, disease extent, and perhaps response to therapy as independent risk factors for the development of CRC. At the same time, ongoing genomic and epigenetic studies are underway to elucidate the mechanism(s) driving the increased risk of CRC observed in our patients with UC. The confluence of these efforts will ultimately provide clinicians with the information necessary to identify individual patients at the greatest risk for the development of CRC, and contribute to the development of more effective invasive and noninvasive markers of CRC and more effective chemoprophylactic strategies. Until this level of personalized medicine is within reach, clinicians must decide how and when to include a discussion of CRC when counseling individual patients with UC and their parents, and how to manage CRC risk in their clinical practice. Clinical experience, in conjunction with data from >40 years of aggregate research, demonstrates that the incidence of CRC in children and young adults is quite small. However, the severe nature of this complication mandates its inclusion in any discussions concerning longterm risks associated with UC. Though our present understanding of CRC is incomplete, available data do afford clinicians with the opportunity to begin to consider individual patients along particular demographic and clinical dimensions in efforts to identify those at particular risk for the development of CRC. • How old was the patient when UC was diagnosed? Patients that were diagnosed with UC as children or young adults (age 0–19 years) may have an RR of developing CRC that reaches 43.85. In contrast, the RR falls to 2.65 if patients were diagnosed with UC between ages of 20–39 years and to background if they were diagnosed with UC after the age of 60 years. • How long has the patient carried a diagnosis of UC? The risk of developing CRC appears to be related to the length of time that a patient has been experiencing chronic intestinal inflammation. As such, surveillance strategies should be based on the interval since their diagnosis and not their present chronologic age [1]. • How extensive is the patient’s disease? Multiple studies have demonstrated that the risk of CRC is highest in those patients with pancolitis. Patients with disease
IBD and Neoplasia in Children
Dig Dis 2014;32:455–462 DOI: 10.1159/000358152
457
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
Eaden et al. [4] completed a meta-analysis in an effort to assess the risk of CRC in patients with ulcerative colitis. They were able to identify 26 studies from the adult literature and an additional 5 studies that estimated both the incidence of CRC in children (average age at diagnosis was 10 years) with UC and reported duration of follow-up (average 12 years). They reported a cumulative incidence of CRC for adults diagnosed with UC at 4.4, 8.6, and 12.7% at 10, 20, and 30 years after diagnosis, respectively. Data from the five eligible studies that included children reported similar, albeit higher (5.5, 10.8, and 15.7%) cumulative risk rates for CRC. Interestingly, data from this study also suggested a geographic influence with greater rates of CRC observed in studies from the UK and United States, compared to those from Scandinavia. As noted above, it is unclear if these differences are the results of underlying genetic or dietary covariance, environmental exposures, or medical practice styles. Ekbom et al. [5] completed a population-based study in Sweden. They reviewed the medical charts of 613 patients (97 aged 50 years, respectively). The SIR for CRC also dropped steadily in cohorts diagnosed with UC with increasing age: 60 years (0.12). There is some suggestion that the incidence of CRC in patients with UC may be decreasing over time. It remains speculative as to whether or not this apparent decrease is related to improved surveillance strategies, more aggressive pharmacologic management, or earlier and more effective surgical approaches to the management of patients with UC. These more recent findings support the hypothesis that the translation of any conclusions drawn from any single study data must also take into account the interval being studied. In an especially large and comprehensive population-based study employing the Danish
Skin Cancer in Children with IBD
Skin cancers present a growing health burden in the US and abroad. In general, these cancers are separated into melanoma and nonmelanoma conditions. Melanomas arise from melanin-producing pigment cells found in the basal layer of the epidermis. They can occur in both sun-exposed and nonexposed parts of the body. Melanomas can be very aggressive. They are invasive and can metastasize early and before there is an opportunity for curative resection. Nonmelanoma skin cancers (NMSC) 458
Dig Dis 2014;32:455–462 DOI: 10.1159/000358152
include basal (BCC) and squamous cell carcinomas (SCC) [10]. Both NMSC forms typically arise in sun-exposed areas. BCC are much more common, represent 74% of all new diagnoses, and arise from cells in the lower part of the epidermis. BCC tend to be more superficial, indolent, and amenable to surgical resection. In contrast, SCC are derived from superficial keratinocytes, have a greater tendency to be invasive, and are more likely to metastasize early and recur [11]. Current data indicate that skin cancers are on the rise, and despite intensive marketing of sun-blocking products, the incidence of all forms of skin cancer have increased by 1.9% per year, among both men and woman, in the interval between 2000 and 2009. The mortality related to skin cancer has similarly increased by approximately 1% per year [12]. While gender has a modest impact on the risk of developing skin cancer, there appears to be a much more significant racial impact. While skin cancer rates have climbed in Caucasian populations, incidence and mortality rates have remained relatively stable over the same interval for darker-skinned (black and Hispanic) individuals. Sun exposure drives the development of skin cancer and, as such, geography plays a significant role in the development of these malignancies. With respect to NMSC, the risk appears to be related to the chronicity or cumulative effect of sun exposure [13]. In contrast, the risk of melanoma appears to be much more related to the intensity of intermittent exposures, such as that resulting in blistering sunburns [14]. Sun exposure generally varies with latitude and has been standardized by the World Health Organization in the form of an international UV index. UV indices vary from about 3.5 in Paris and Vancouver and rise to 11.5 in more southern latitudes, including Bangkok (10.5) and Singapore (11.5) [15]. The concerns about the possibility of an increased risk of skin cancer in patients with IBD, especially those receiving immunosuppressive agents, arose from earlier reports in the transplant literature. Solid organ transplant recipients, especially those receiving the same immunomodulatory agents now frequently used in the long-term management of both Crohn’s disease (CD) and UC, appear to be especially prone to the development of NMSC, and the malignancies in these patients tend to occur earlier and are more aggressive [16]. The risk of SCC is especially high and may be up to 65 times greater than in the general population [17]. The mechanism(s) underlying the increased risk of skin cancer in patients receiving immunosuppressive therapy remain incompletely understood. Existing in vitro data demonstrate that thiopurine Rufo
Downloaded by: Kungliga Tekniska Hogskolan 198.143.54.65 - 8/13/2015 11:44:13 PM
primarily on the left side of their colon are at intermediate risk. In most studies, patients with limited proctitis appear to be at no increased risk for the development of CRC than age-matched peers [1]. It remains unclear if the presence of backwash ileitis is an independent risk factor for the development of CRC [7]. • Does the patient have PSC? Population-based studies indicated that patients with PSC and IBD are at considerably greater risk (RR: 9.13) compared to patients with UC and no PSC [6]. A meta-analysis similarly supported this correlation and reported an odds ratio (OR) of 4.79 [8]. Data further suggest that patients with PSC are at particular risk for the early development of CRC [9]. For unclear reasons, CRC in patients with PSC is more likely to occur on the right side of the colon. In contrast, right-sided CRC is noted in only 38% of patients with IBD and no PSC. As such, current guidelines indicate that annual full surveillance colonoscopic evaluations should be performed in all patients with UC and PSC within 2 years of diagnosis. • Does the patient have a family history of CRC? Existing data suggest that a genetic context can serve as an additional risk modifier for the development of CRC in patients with UC. If a patient with UC has a first-degree relative with CRC, their RR of CRC doubles. The risk increases 9-fold if the family member of the UC patient was
