A review of the literature revealed significant associations between 23 different autoimmune and chronic inflammatory autoimmune-related diseases and cancer ( and ). Of these observed relationships, celiac disease had some of the strongest and most extensive associations with both focal organ and lymphoproliferative malignancies. Celiac disease is a complex condition in which immune intolerance for naturally ingested wheat protein, gluten, and antibodies towards tissue transglutaminases lead to the development of significant enteropathy and associated symptoms (). Patients who are positive for human lymphocyte antigen classes DQ2 or DQ8 are particularly affected by increased binding of antigen-presenting cells to deamidated gluten, allowing for induction of a CD4+
helper T-cell-mediated inflammatory response (24
). This response includes IFN-γ release and CD8+
T-cell activation, which leads to tissue damage and gut cytotoxicity, creating a pro-malignancy chronic inflammatory state. One group evaluating this connection was that of Askling et al.
, who used Swedish National Registers to follow a cohort of approximately 11,000 patients with celiac disease to evaluate cancer incidence over an average of 9.8 years (25
). This group calculated a standardized incidence ratio (SIR) of 10 for small bowel malignancy, 4.2 for esophageal cancer, and 6.3 for non-Hodgkin's lymphoma (NHL) among other significant associations (25
). Several other studies confirmed this association of celiac disease and NHL, with calculations ranging from an adjusted rate ratio (ARR) of 3.28 (26
) to a standardized morbidity ratio (SMR) of 9.1 (27
). T-cell NHL, specifically, was associated with celiac disease, with an odds ratio (OR) of 5.9 (28
), a relative risk (RR) of 17 (29
) and an SIR of 51 (30
Reported associations between autoimmune diseases, chronic inflammatory diseases, and focal malignancies.
Reported associations between autoimmune diseases, chronic inflammatory diseases, and hematologic malignancies.
Figure 3 Celiac disease as a paradigm for chronic inflammation predisposing patients to cancer development. Autoimmune processes and inflammation believed to contribute to cancer are shown (24, 63, 64). HLA, human lymphocyte antigen; Th1, helper T-cell type 1; (more ...)
Other conditions demonstrating extensive associations with cancer include Crohn's disease and ulcerative colitis, collectively referred to as inflammatory bowel disease (IBD). IBD, like celiac disease, exhibited associations with gut malignancies, the primary system targeted by its inflammation. In fact, the classic pathology of IBD – granulomatous and transmural gut mucosal inflammation in Crohn's disease and primarily distal mucosal and submucosal inflammation in ulcerative colitis – may be a result of a unique milieu of inflammatory mediators experienced in each condition. This abnormal immune response appears to be the result of multifaceted immune dysregulation. For example, genetic mutations in patients with IBD have been isolated and linked to aberrancies in autophagy and eradication of antigenic material, subsequently causing abnormally high levels of inflammation. Moreover, unlike the typical immunosuppressive gut environment of unaffected people, patients with IBD appear to lose immune tolerance for normal gut flora, again creating an environment of chronic inflammation. This loss of tolerance extends to the adaptive immune systems of patients with IBD, which appear to be sensitized to self-antigens and release interleukins (IL) and interferons that eventually lead to local tissue damage. Furthermore, continued lack of immune tolerance is likely perpetuated through defects in Treg function. While Tregs typically secrete IL-10, suppressing a robust immune response, human and mouse models suggest that IBD-like pathology can develop with abnormal Tregs and impaired immune tolerance (31
). Resultant chronic inflammation is likely the source of increased cancer risk observed in patients with IBD. In one study, Hemminki et al.
evaluated this relationship by following a Swedish cohort of nearly 22,000 patients with Crohn's disease for increased development of malignancy (33
). This population demonstrated an SIR for cancer in general of 1.54; higher associations were calculated for small bowel carcinoma at 13.82, colon cancer at 2.93, and NHL at 2.54 (33
). Ulcerative colitis, likewise, carried an elevated cancer risk based on another Swedish cohort study of approximately 28,000 patients (general cancer SIR 1.46, small bowel 2.42, colon 3.60, NHL 1.52) (34
). Other studies described that the longitudinal risk of cancer and/or dysplasia in IBD increased proportionally to the number of years post-diagnosis, and was as high as 50% at 25 years (35
Strong and recurrent associations were not limited to diseases which focally target one organ or organ system. Autoimmune conditions such as rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE), which affect multiple organ systems, had significant associations with both the development of focal types of cancer throughout the body and lymphoproliferative and hematologic malignancies. RA is an autoimmune disease that primarily affects joints and cartilage through the development of pannus, a product of inflammatory cells and mediators that transform synoviocytes within patients' joints to become locally invasive and destructive. The exact etiology of this condition remains to be fully characterized, as RA can also cause systemic manifestations affecting the skin, vasculature, heart, lungs, and peripheral nerves (37
). RA demonstrated many significant associations with various types of malignancies. In one study, Chen et al.
followed a Taiwanese cohort of over 23,000 patients with RA for cancer development for an average of 5.90 years (39
). Results demonstrated mildly, yet significantly elevated SIR for cancer in patients with RA. These associations included an overall association with cancer at 1.12, lung cancer at 1.36, kidney cancer at 2.12, thyroid cancer at 1.41, melanoma at 1.47, all hematologic cancer at 2.74, NHL at 3.54, Hodgkin's lymphoma at 1.76, and all leukemia at 1.48 (39
). Other studies also show a broad range of cancer types associated with RA. One group, Ekstrom et al.
, utilized a Swedish database and specifically evaluated the incidence of hematologic cancer in over 75,000 patients with RA (40
). Their results demonstrated significant SIR for any type of hematologic cancer at 1.07, any type of lymphoma at 2.00, NHL at 1.89, and Hodgkin's lymphoma at 3.06 (40
Like RA, SLE has a complex multifactorial genetic and environmental etiology affecting multiple organ systems resulting in widespread loss of immune self-tolerance (41
), and also showing evidence of increased cancer risk. In another Taiwanese cohort study, Chen et al.
followed nearly 12,000 patients with SLE for cancer incidence. This study established multiple positive associations between SLE and cancer: the highest SIRs were for hematologic cancer at 4.96, NHL at 7.27, and leukemia at 2.64 (42
). This study also demonstrated increased SIR for solid tumors; the risk was particularly elevated in the central nervous system at 3.30, and the kidney at 3.99 (42
). Another study demonstrating multiple associations between SLE and cancer was performed by Parikh-Patel et al.
on a Californian database of over 30,000 patients with SLE. This group was followed for an average of 5.1 years, and within that time period, associations for multiple types of cancer were identified: the SIR include 1.14 for any type of cancer, 1.66 for lung cancer, 2.15 for kidney cancer, 2.74 for NHL, 3.02 for Hodgkin's lymphoma, 2.13 for any type of leukemia, 2.96 for myelogenous leukemia, 3.26 for diffuse large B-cell lymphoma, and 2.89 for follicular lymphoma (43
Unlike the majority of autoimmune and chronic inflammatory conditions investigated, one condition was inversely associated with cancer development in most organ systems: multiple sclerosis (MS). MS is a neurodegenerative condition where T-cell-mediated, focal autoimmune attacks against the myelin sheath surrounding nerves leads to the development of demyelinated neuronal plaques and nervous dysfunction (44
). Hazard ratios (HR) were calculated in one study by Bahmanyar et al.
of over 20,000 Swedish patients with MS to quantify the likelihood of cancer development. After an average observation period of 35.1 years, results were most notable for a decreased overall HR for cancer of 0.91, lung cancer of 0.69, ovarian cancer of 0.58, cervical cancer of 0.58, prostate cancer of 0.80, any gastrointestinal malignancy of 0.83, pancreatic cancer of 0.67, and lymphoma of 0.76 (45
). This same study, however, showed an increased HR for brain neoplasms at 1.44 (45
). This result is reiterated in another study by Sumelahti et al.
investigating a smaller Finnish cohort of approximately 1,600 patients with MS evaluated to have an SIR of central nervous system malignancies of 1.3; overall association with cancer was not significant (46
). The decrease in risk observed in other sites was hypothesized to be related to immunomodulatory treatment regimens or a hypothetical increase in immune surveillance concurrent with the ongoing heightened immune activity in the disease (45
). The decrease in cancer risk does not appear to be conferred to first-degree relatives, and in a disease such as MS which is highly heritable, a genetic basis for the decreased risk is less likely (45
Although other incidences were not as pervasive as the relationships of MS with types of cancer, other conditions also demonstrated occasional risk calculations suggesting a diminished risk of cancer. In the study described above by Chen et al.
evaluating RA, the reported SIR of cervical cancer was 0.86, colorectal cancer was 0.94, and non-melanoma skin cancer was 0.87 (39
). In the study by Parikh-Patel et al.
, SLE also exhibited an SIR of 0.55 for cervical cancer, 0.76 for breast cancer, 0.69 for prostate cancer, and 0.67 for melanoma (43
). Such observations may be linked to increased focal immunosurveillance in these conditions, although exact mechanisms remain unclear. Parikh-Patel et al.
suggest that these decreases in focal risk are unlikely to be a result of increased preventative screening measures in patients with SLE, as this essential aspect of care is frequently neglected (43
Generally, literature suggested that patients with autoimmune and chronic inflammatory disease are at higher risk for cancer development; however, as shown in , many autoimmune conditions were primarily diagnosed in patients after an initial detection of cancer. Frequent publication of case reports documenting these relationships exist, although risk calculations are largely unavailable. Certain conditions, such as myasthenia gravis
or Lambert-Eaton myasthenic syndrome (LEMS), were so frequently observed to occur within the clinical course of malignancy that their management includes an investigation for tumors (47
). Of the total patient population diagnosed with small cell lung cancer (SCLC), 3% show signs of LEMS (51
), which is characterized by antibodies targeting pre-synaptic calcium channels at the neuromuscular junction, but only the minority (30–50%) of diagnosed cases of LEMS will not eventually be associated with an established SCLC (48
). This strong correlation has led patients with LEMS to be recommended to undergo biyearly screening to ensure a lung malignancy has not been overlooked (48
). Other autoimmune conditions, such as autoimmune hemolytic anemia, may occur in a broader range of cancer types, and may be detected prior to a diagnosis of or within the course of a lymphoproliferative malignancy (28
), notably appearing in both NHL (57
) and in multiple myeloma (59
). All conditions included in have multiple reports, suggesting the condition occurs in the setting of a pre-existing cancer, as opposed to a temporally coincidental onset of autoimmunity with cancer.
Cancer types frequently associated with paraneoplastic autoimmunity.