Abstract
Background:
Autoimmune diseases and hematopoietic malignancies are known to cluster within individuals, suggesting intertwined etiologies. A limited number of studies have evaluated pre-existing medical conditions as risk factors for myelodysplastic syndromes (MDS). We evaluated associations between autoimmune disease and other medical conditions and risk of MDS.
Methods:
Cases were identified through the Minnesota Cancer Reporting System. Controls were identified through the Minnesota State driver’s license/identification card list. History of autoimmune disease and other medical conditions was based on self-report; proxy interviews were not conducted. Unconditional logistic regression was used to calculate adjusted odds ratios (aORs) and 95% confidence intervals (CI).
Results:
We included 395 cases and 694 controls. Cases were significantly more likely to report a diagnosis of any autoimmune disease when compared with controls (aOR=1.41, 95% CI: 1.05–1.89) after adjustment for age, sex, education, NSAID use, exposure to benzene and body mass index. When we evaluated specific autoimmune conditions, a statistically significant association was observed for hypothyroidism (aOR=2.16, 95% CI: 1.39–3.34) and odds ratios were elevated for inflammatory bowel disease (aOR=1.75) and systemic lupus erythematosus (SLE; aOR=3.65), although these associations did not reach statistical significance. Presence of an autoimmune condition did not impact overall survival (p=0.91).
Conclusion:
Our results validate previous findings of an association between autoimmune disease and MDS. Further studies are required to determine whether this association is due to shared etiology, treatment for autoimmune diseases, or altered immune surveillance or bone marrow damage caused by the autoimmune condition.
Keywords: Myelodysplastic syndrome, autoimmune disease, chronic immune stimulation, risk factor
Introduction
Myelodysplastic syndromes (MDS) are a heterogeneous collection of neoplastic conditions that result in dysplastic and ineffective hematopoiesis and that increase in incidence throughout the lifespan.[1] Accordingly, the overall age-adjusted incidence rate in the US was 4.8 per 100,000 person-years as reported by the National Cancer Institute’s Surveillance, Epidemiology and End Results Program (SEER 18), with a maximum rate of 55.5 per 100,000 observed in those 80 years of age and older.[2] Interestingly, rates in Minnesota were the third highest reported rates in the country behind Washington state and Idaho, making this an ideal population for etiologic study.[3]
Little is known about the etiology of MDS, although clinical studies have shown that chemotherapy (particularly alkylating agents) or radiation therapy from a previous cancer can cause MDS.[4–7] These therapy-related MDS cases account for a small percentage of diagnoses (5%)[3, 8, 9] and have a particularly dismal prognosis, while the majority of cases are de novo. Given that ~1/3 of MDS cases eventually develop AML,[10] it is likely that the two diseases have overlapping etiologies. Known risk factors for AML include increased age, male sex, cigarette smoking, obesity, exposure to benzene and other chemicals, and prior chemotherapy.[11–13] Evidence also exists to support these exposures as risk factors for MDS.[12, 14–21]
In addition, medical conditions have been evaluated as risk factors for MDS, with evidence that personal history of autoimmune disease is associated with risk.[22, 23] Data from the Swedish population-based central registries[23] show that a personal history of any autoimmune disease is associated with an increased risk for both AML (OR=1.7, 95% CI 1.5, 1.9) and MDS (OR=2.1, 95% CI 1.7, 2.6) with stronger associations observed for individual autoimmune diseases. Similarly, using data from SEER and Medicare, Anderson et al.[22] identified an increased risk for MDS/AML associated with autoimmune diseases, although the risk estimates were more modest. Several case-control studies have also suggested an association between autoimmune disease and MDS.[24, 25] This increased risk following a diagnosis for an autoimmune condition is hypothesized to be due to shared genetic or environmental risk factors, treatment for autoimmune diseases, or damage to the bone marrow by the condition.[22]
In this Minnesota-based case-control study, we examined the association between self-reported medical conditions, including autoimmune diseases, and risk of MDS. Given the prior reports of patient medical histories,[22–24, 26] we hypothesized that cases would more often have a diagnosis of autoimmune and other chronic conditions predating their MDS diagnosis compared to unaffected controls.
Methods
Data collection procedures for the Adults in Minnesota with Myelodysplastic Syndrome (AIMMS) Study have been previously described.[14, 15] Cases and controls were eligible for the study if they were Minnesota residents between the ages of 20 and 85 years and could understand English or Spanish. Proxy interviews were not conducted.
MDS Cases
Cases were identified by the Minnesota Cancer Reporting System (MCRS), a population-based registry that collects information on all cancers diagnosed in Minnesota. Due to the rapidly fatal nature of MDS, a rapid case ascertainment system was used to identify cases. Pathology logs were typically received and reviewed by MCRS staff within 1–2 months of diagnosis. Cases were eligible for the study if they had a diagnosis of MDS (ICD-O-3 codes: 9980, 9982–9987, 9989) between April 1, 2010 and October 31, 2014. Centralized pathology and cytogenetics reviews were conducted to confirm diagnoses and classify subtypes. Only cases with confirmed MDS following independent reviews by two board-certified hematopathologists, a cytogeneticist, and a medical oncologist were included in the analysis. MDS subtypes were categorized based on WHO classification using the 2008 revised criteria.[27] MDS cases were classified as de novo MDS (all subtypes other than therapy-related) or therapy-related MDS based on the consensus diagnosis from the central review process, which included review of diagnostic specimens, pathology reports, flow cytometry and medical record review and was not based solely on history of cancer or prior treatment.
Controls
Controls were identified through the Minnesota State driver’s license/identification card list which includes virtually all adults living in Minnesota.[28] Controls were eligible if they were alive at the time of contact, resided in Minnesota, were between the ages of 20 and 85 years, could understand English or Spanish, and had no prior diagnosis of myeloid malignancy. Controls were frequency matched to cases on sex and decile of age.
Medical History Data
Exposure data were collected by a self-administered questionnaire that queried demographics, anthropometrics, lifestyle factors, physical activity, medication use, medical history, reproductive history, family cancer history, farm/rural living, pesticide exposure, occupational exposures, and residential chemical exposures. Assessment of medical history was focused on a time point two years prior to questionnaire completion to allow for a potential latency period for MDS development. The medical history section of the questionnaire included items regarding regular use of NSAIDs, height, weight at various time points during adulthood, blood transfusions, transplants, history of medical conditions and history of cancer. Study participants were asked if they had been diagnosed with the following health conditions by a doctor or health professional: hypothyroidism, rheumatoid arthritis, hyperthyroidism, ulcerative colitis, polymyositis, dermatomyositis, or polymyalgia rheumatica, diabetes mellitus, Crohn’s disease, Lupus or systemic lupus erythematosus, celiac disease, ankylosing spondylitis, Sjogren’s disease or sicca syndrome, high blood pressure, heart disease, angina, or heart attack, osteoarthritis, eczema, infectious mononucleosis, peptic ulcer, contact dermatitis, chronic fatigue syndrome, cirrhosis of the liver or liver damage, and epilepsy. Individuals who reported a diagnosis were asked to report age at first diagnosis. Participants were also asked to report any previous diagnoses of cancer, including cancer type, diagnosis age, and treatment type (surgery, radiation, chemotherapy, none, other, or don’t know).
Protection of Human Subjects
The Institutional Review Boards for each participating institution approved this study, which included the University of Minnesota, the Mayo Clinic, the Minnesota Department of Health, and area hospitals. Informed consent was obtained from all participants.
Statistical Analysis
Multivariable unconditional logistic regression was used to quantify associations between autoimmune and other chronic diseases and MDS; adjusted odds ratios (aORs) and 95% confidence intervals (CIs) were computed. For outcomes, we examined associations for MDS overall and also conducted analyses stratified by sex. An interaction term was included in the model to determine whether the association between autoimmune disease and MDS differed in males and females. In order to examine the effect of different latency periods, we excluded cases where the autoimmune disease was diagnosed less than 2 or less than 5 years prior to diagnosis of MDS. Polytomous logistic regression was used to evaluate the association between MDS type (de novo or therapy related) and autoimmune and other chronic diseases.
The exposures, individual autoimmune and chronic conditions, were coded as dichotomous variables (i.e., present vs. absent). Age (continuous) and sex, the frequency matching variables, were included in all models as a matter of course. Other variables considered as potential confounders were selected based on previous reported associations in the literature and included race/ethnicity (non-Hispanic white, other), education (≤high school graduate, some post-high school training, college graduate), household income (≤$40,000, >$40,000 - $80,000, >$80,000), smoking status (never, former, current), occupational and recreational physical activity, use of nonsteroidal anti-inflammatory drugs (NSAID) at least once per week for at least one year (yes, no), personal and family history of cancer (yes, no), prior cancer treatment (yes, no), and prior exposure to benzene (yes, no). Potential confounders were included in the final models if they changed the parameter estimate for the association between autoimmune disease and MDS by more than 10%.
We also ran a case-only analysis using the methods described above to compare the distributions of sex, age category, MDS subtype, IPSS-R risk category and cytogenetics by presence/absence of autoimmune disease.
We used standard survival analysis techniques to examine potential differences in overall survival in cases with and without autoimmune diseases. Kaplan-Meier survival curves were constructed and median survival times for these groups were compared via log-rank tests. A multivariable Cox proportional hazards model with adjustment for age, sex, education, BMI, NSAID use and benzene exposure was used to compute the adjusted hazard ratio (aHR) for death in cases with autoimmune disease after confirming the validity of the proportional hazards assumption in this dataset.
All analyses were performed using SAS (Version 9.4, SAS Institute Inc., Cary, NC, USA) and all reported p-values were two-sided. Odds ratios were not calculated when there were fewer than five exposed cases in a category.
Results
Response rates and subject characteristics
During the study eligibility period, MCRS screened 1,031 potentially eligible MDS cases for study referral. Of the 881 potentially eligible and contacted cases, 558 completed questionnaires and 564 provided written informed consent for medical record review for (63% response rate). Using the central review process, our team confirmed an eligible diagnosis of MDS in 399 (71%) of the potential cases. Recruitment of cases was conducted from 4/2010–3/2015. The median time from diagnosis to questionnaire completion was 140 days (range: 7–1333). A total of 698 controls were recruited (response rate 49%). Controls were recruited from 7/2010–7/2014 and the median time from first contact to questionnaire completion was 39 days (2–1130 days).
Four cases and four controls did not have complete data for the medical history and/or adjustment variables and were excluded from the analyses. As shown in Table 1, cases were largely similar to controls, except cases were more likely to be age 80 years or older and more likely to report prior exposure to benzene. The vast majority of participants reported white race and non-Hispanic ethnicity. The distribution of BMI, education, and NSAID use was similar in the case and control groups.
Table 1.
Descriptive characteristics of the study population.
| Controls N = 694 |
Cases N = 395 |
|
|---|---|---|
| N (%) | N (%) | |
| Sex | ||
| Male | 438 (63) | 254 (64) |
| Female | 256 (37) | 141 (36) |
| Age at diagnosis (years) | ||
| <50 | 16 (2.3) | 10 (2.5) |
| 50 – 59 | 105 (15.1) | 36 (9.1) |
| 60 – 69 | 201 (28.9) | 106 (26.8) |
| 70 – 79 | 282 (40.6) | 151 (38.2) |
| ≥80 | 90 (13.0) | 92 (23.3) |
| Hispanic/Latino | 9 (1.3) | 1 (0.3) |
| No | 685 (98.7) | 394 (99.8) |
| Race | ||
| White | 681 (98.1) | 387 (98.0) |
| Other | 13 (1.8) | 8 (2.0) |
| Education | ||
| ≤HS grad | 202 (29.1) | 145 (36.7) |
| Some post-HS | 226 (32.6) | 108 (27.3) |
| College grad | 266 (38.3) | 142 (35.9) |
| BMI, 2 years ago (kg/m 2 ) | ||
| <25.0 | 171 (24.6) | 107 (27.1) |
| 25.0 – 29.9 | 291 (41.9) | 148 (37.5) |
| ≥30.0 | 232 (33.4) | 140 (35.4) |
| NSAID Use | ||
| Yes | 513 (73.9) | 289 (73.2) |
| No | 181 (26.1) | 106 (26.8) |
| Benzene exposure | ||
| Yes | 80 (11.5) | 64 (16.2) |
| No/not sure | 614 (88.5) | 331 (83.8) |
aOR: adjusted odds ratio; BMI: body mass index; CI: Confidence Interval; HS: high school. Numbers may not sum to total due to missing values.
Odds ratios not computed for matching variables (age and sex), race and ethnicity.
ORs adjusted for age (continuous), sex, education, BMI, NSAID use and benzene exposure.
Autoimmune diseases
Overall, 30% of cases and 23% of controls reported a prior diagnosis of an autoimmune disease, corresponding to an adjusted OR (aOR) of 1.41 (95% CI: 1.05–1.89) (Table 2). This association was principally attributable to a prior history of hypothyroidism (aOR=2.16, 95% CI: 1.39–3.34), which was reported by 13% of cases and 7% of controls. We also observed elevated odds ratios for less common diagnoses, including inflammatory bowel disease (aOR=1.75, 95% CI: 0.89–3.42), SLE (aOR=3.65, 95% CI: 1.01–13.2), and polymyositis (aOR=2.22, 95% CI 0.87–5.64) although some of these associations did not reach statistical significance. We collected information on several additional conditions but we were not able to calculate odds ratios due to the small number of cases with each condition (early onset diabetes mellitus (N=2), celiac disease (N=3), ankylosing spondylitis (N=5), Sjögren’s disease/Sicca syndrome (N=2)). When we excluded cases where the autoimmune disease was diagnosed less than 2 or less than 5 years prior to diagnosis of MDS, respectively, we observed similar associations to those reported overall (Supplementary Table 1).
Table 2.
Association between personal history of autoimmune and other chronic diseases and MDS.
| Controls N = 694 |
Cases N = 395 |
OR (95% CI) | p-value | |||
|---|---|---|---|---|---|---|
| N (%) | N (%) | aOR1 (95% CI) | p-value | |||
| Autoimmune diseases | ||||||
| Any autoimmune disease | 157 (22.6) | 117 (29.6) | 1.44 (1.08–1.92) | 0.01 | 1.41 (1.05–1.89) | 0.02 |
| None | 537 (77.4) | 278 (70.4) | Ref | |||
| Hypothyroidism | 47 (6.7) | 50 (12.7) | 2.08 (1.35–3.20) | 0.0009 | 2.16 (1.39–3.34) | 0.0006 |
| None | 647 (93.2) | 343 (87.3) | Ref | |||
| Rheumatoid arthritis | 60 (8.6) | 32 (8.1) | 0.92 (0.59–1.44) | 0.72 | 0.80 (0.50–1.26) | 0.33 |
| None | 634 (91.3) | 362 (91.9) | Ref | |||
| Hyperthyroidism | 36 (5.2) | 23 (5.9) | 1.04 (0.60–1.80) | 0.89 | 1.06 (0.61–1.84) | 0.85 |
| None | 658 (94.8) | 370 (94.2) | Ref | |||
| IBD | 19 (2.7) | 18 (4.6) | 1.68 (0.87–3.26) | 0.12 | 1.75 (0.89–3.42) | 0.10 |
| None | 674 (97.3) | 376 (95.4) | ||||
| Polymyositis, dermatomyositis, or polymyalgia rheumatica | 8 (1.1) | 11 (2.8) | 2.09 (0.83–5.29) | 0.12 | 2.22 (0.87–5.64) | 0.09 |
| None | 684 (98.8) | 383 (97.2) | Ref | |||
| Systemic lupus erythematosus | 4 (0.6) | 6 (1.5) | 3.16 (0.87–11.5) | 0.08 | 3.65 (1.01–13.2) | 0.05 |
| None | 688 (99.4) | 389 (98.5) | Ref | |||
| Other chronic diseases | ||||||
| High blood pressure | 376 (54.2) | 223 (56.6) | 0.96 (0.74–1.24) | 0.76 | 0.97 (0.74–1.27) | 0.80 |
| None | 318 (45.8) | 171 (43.4) | Ref | |||
| Heart disease, angina or heart attack | 143 (20.6) | 128 (32.5) | 1.66 (1.24–2.21) | 0.0006 | 1.71 (1.27–2.31) | 0.0005 |
| None | 550 (79.4) | 266 (67.5) | Ref | |||
| Osteoarthritis | 141 (20.3) | 79 (20.1) | 0.91 (0.67–1.25) | 0.57 | 0.93 (0.68–1.29) | 0.67 |
| None | 552 (79.7) | 315 (80.0) | Ref | |||
| Diabetes mellitus (>30 y) | 113 (16.3) | 75 (19.0) | 1.20 (0.87–1.66) | 0.27 | 1.19 (0.84–1.67) | 0.32 |
| None | 581 (83.7) | 319 (81.0) | Ref | |||
| Eczema | 58 (8.4) | 33 (8.4) | 0.98 (0.63–1.54) | 0.94 | 1.00 (0.63–1.57) | 0.99 |
| None | 633 (91.6) | 362 (91.7) | Ref | |||
| Infectious mononucleosis | 43 (6.2) | 24 (6.1) | 1.19 (0.71–2.00) | 0.52 | 1.17 (0.69–1.99) | 0.55 |
| None | 649 (93.8) | 371 (93.9) | Ref | |||
| Peptic ulcer | 44 (6.3) | 28 (7.1) | 1.03 (0.63–1.69) | 0.90 | 1.02 (0.62–1.69) | 0.93 |
| None | 649 (93.7) | 366 (92.9) | Ref | |||
| Contact dermatitis | 24 (3.5) | 18 (4.6) | 1.38 (0.74–2.57) | 0.31 | 1.21 (0.64–2.30) | 0.55 |
| None | 668 (96.5) | 376 (95.4) | Ref | |||
| Chronic fatigue syndrome | 6 (0.9) | 13 (3.3) | 3.63 (1.36–9.71) | 0.01 | 4.07 (1.51–11.0) | 0.006 |
| None | 686 (99.1) | 382 (96.7) | Ref | |||
| Cirrhosis of the liver or liver damage | 12 (1.7) | 9 (2.3) | 1.38 (0.57–3.32) | 0.47 | 1.47 (0.61–3.58) | 0.39 |
| None | 680 (98.3) | 386 (97.7) | Ref |
aOR = adjusted odds ratio; BMI: body mass index; CI = confidence interval; HS: high school; MDS: myelodysplastic syndrome; NSAID: nonsteroidal anti-inflammatory drug.
Numbers in tables may not sum to total number of cases/controls due to missing values.
ORs adjusted for age (continuous), sex, education (≤ HS grad, some post HS, college grad), NSAID use, exposure to benzene, BMI 2 years ago (< 25.0, 25.0–29.9, ≥ 30.0 kg/m2)
When we stratified the analysis by sex, the association between any autoimmune disease and MDS was statistically significant only in females, while the association was attenuated in males although this difference did not reach statistical significance (p interaction = 0.09; Supplementary Table 2). As expected, autoimmune diseases were more common in female cases and controls as compared to their male counterparts, 45% vs. 21%. When we evaluated individual autoimmune conditions, we observed a significant association between hypothyroidism and MDS in both males and females. We also observed similar aOR for other autoimmune conditions with sufficient sample size to permit calculation of aORs in both groups.
When we evaluated risks separately for de novo versus therapy-related MDS, the association between any autoimmune disease and MDS was borderline statistically significant in both the therapy-related group (aOR=1.94, 95% CI: 0.95–3.94) and the de novo MDS group (aOR=1.35, 95% CI: 0.99–1.83; Table 3). We observed a strong, statistically significant positive association between hypothyroidism and both therapy-related and de novo MDS. In the de novo group, we also observed elevated odds ratios for associations between MDS and inflammatory bowel disease and polymyositis. We did not have sufficient sample sizes to calculate aOR for the remaining conditions.
Table 3.
Association between a personal history of autoimmune and other chronic diseases and de novo and therapy-related MDS.
| Controls N = 694 |
De novo MDS N = 356 |
Therapy-related MDS N = 39 |
|||||
|---|---|---|---|---|---|---|---|
| N (%) | N (%) | aOR1 (95% CI) | p-value | N (%) | aOR1 (95% CI) | p-value | |
| Autoimmune diseases | |||||||
| Any autoimmune disease | 272 (19.7) | 103 (28.9) | 1.35 (0.99–1.83) | 0.06 | 14 (35.9) | 1.94 (0.95–3.94) | 0.07 |
| None | 1110 (80.3) | 253 (71.1) | Ref | 25 (64.1) | Ref | ||
| Hypothyroidism | 47 (6.8) | 42 (11.9) | 1.97 (1.25–3.12) | 0.004 | 8 (20.5) | 3.44 (1.42–8.32) | 0.006 |
| None | 647 (93.2) | 312 (88.1) | Ref | 31 (79.5) | Ref | ||
| Rheumatoid arthritis | 60 (8.7) | 29 (8.2) | 0.78 (0.49–1.26) | 0.31 | 3 (7.7) | - | - |
| None | 634 (91.4) | 326 (91.8) | Ref | 36 (92.3) | |||
| Hyperthyroidism | 36 (5.2) | 22 (6.2) | 1.10 (0.62–1.95) | 0.74 | 1 (2.6) | - | - |
| None | 658 (94.8) | 332 (93.8) | Ref | 38 (97.4) | |||
| Inflammatory Bowel Disease | 19 (2.7) | 16 (4.5) | 1.75 (0.87–3.50) | 0.11 | 2 (5.1) | - | - |
| 674 (97.3) | 339 (95.4) | Ref | 37 (94.9) | ||||
| Polymyositis, dermatomyositis, or polymyalgia rheumatica | 8 (1.1) | 11 (3.1) | 2.38 (0.93–6.09) | 0.07 | - | - | - |
| None | 684 (98.8) | 344 (96.9) | Ref | 39 (100.0) | |||
| Other chronic diseases | |||||||
| High blood pressure | 376 (54.1) | 206 (58.0) | 0.99 (0.74–1.31) | 0.91 | 17 (43.6) | 0.82 (0.40–1.68) | 0.59 |
| None | 318 (45.8) | 149 (42.0) | Ref | 22 (56.4) | Ref | ||
| Heart disease, angina or heart attack | 143 (20.6) | 121 (34.1) | 1.74 (1.28–2.37) | 0.0004 | 7 (18.0) | 1.17 (0.48–2.85) | 0.74 |
| None | 550 (79.3) | 234 (65.9) | Ref | 32 (82.1) | Ref | ||
| Osteoarthritis | 141 (20.3) | 71 (20.0) | 0.92 (0.66–1.28) | 0.61 | 8 (20.5) | 1.10 (0.48–2.51) | 0.82 |
| None | 552 (79.7) | 284 (80.0) | Ref | 31 (79.5) | Ref | ||
| Diabetes mellitus (>30 y) | 113 (16.3) | 69 (19.4) | 1.20 (0.84–1.70) | 0.32 | 6 (15.4) | 1.113 (0.44–2.92) | 0.80 |
| None | 581 (83.7) | 286 (80.6) | Ref | 33 (84.6 | Ref | ||
| Eczema | 58 (8.4) | 33 (9.3) | 1.13 (0.72–1.79) | 0.57 | - | - | - |
| None | 633 (91.6) | 323 (90.7) | Ref | 39 (100.0) | |||
| Infectious mononucleosis | 43 (6.2) | 18 (5.1) | 1.01 (0.56–1.80) | 0.98 | 6 (15.4) | 2.23 (0.86–5.77) | 0.10 |
| None | 649 (93.8) | 338 (94.9) | Ref | 33 (84.6) | Ref | ||
| Peptic ulcer | 44 (6.4) | 24 (6.8) | 0.93 (0.55–1.58) | 0.79 | 4 (10.3) | 2.13 (0.70–6.44) | 0.18 |
| None | 649 (93.6) | 331 (93.2) | Ref | 35 (89.7) | Ref | ||
| Contact dermatitis | 24 (3.5) | 18 (5.1) | 1.36 (0.71–2.58) | 0.36 | - | - | - |
| None | 668 (96.5) | 338 (94.9) | Ref | 38 (100.0) | |||
| Cirrhosis of the liver | 12 (1.7) | 8 (2.25) | 1.47 (0.58–3.80) | 0.41 | 1 (2.6) | ||
| None | 680 (98.2) | 348 (97.8) | Ref | 38 (97.4) | |||
| Chronic fatigue syndrome | 6 (0.9) | 12 (3.4) | 4.12 (1.49–11.4) | 0.006 | 1 (2.6) | - | - |
| None | 686 (99.1) | 344 (96.6) | Ref | 38 (97.4) | |||
aOR = adjusted odds ratio; BMI: body mass index; CI = confidence interval; HS: high school; MDS: myelodysplastic syndrome; NSAID: nonsteroidal anti-inflammatory drug.
Numbers in tables may not sum to total number of cases/controls due to missing values.
ORs adjusted for age (continuous), sex, education (≤ HS grad, some post HS, college grad), NSAID use, exposure to benzene, BMI 2 years ago (< 25.0, 25.0–29.9, ≥ 30.0)
In a case-only analysis, females were overrepresented among the cases with autoimmune disease (aOR=3.25, 95% CI: 2.01–5.24; Table 4), with close to half of the female MDS cases reporting an autoimmune disease. We did not observe any significant differences between age at diagnosis, cytogenetics, WHO disease subtype, or IPSS-R category in cases with and without a history of autoimmune disease.
Table 4.
Comparison of disease characteristics in cases with and without a history of autoimmune disease.
| Characteristic | Any Autoimmune Disease N = 117 |
No Autoimmune Disease N = 278 |
|
|---|---|---|---|
| N (%) | N (%) | aOR1 (95% CI) | |
| Sex | |||
| Female | 64 (54.7) | 77 (27.7) | 3.25 (2.01–5.24) |
| Male | 53 (45.3) | 201 (72.3) | Ref |
| Age at diagnosis (years) | |||
| ≤50 | 2 (2.5) | 7 (2.5) | 1.32 (0.29–5.99) |
| 50 – 59 | 11 (9.4) | 25 (9.0) | 1.01 (0.42–2.41) |
| 60 – 69 | 30 (25.6) | 76 (27.3) | Ref |
| 70 – 79 | 52 (44.4) | 99 (35.6) | 1.29 (0.72–2.31) |
| ≥80 | 21 (17.9) | 71 (25.5) | 0.75 (0.38–1.48) |
| Cytogenetics | |||
| Abnormal | 67 (58.3) | 139 (53.8) | 1.05 (0.65–1.68) |
| Normal | 48 (41.7) | 120 (46.2) | Ref |
| WHO Disease Subtype (MDS) | |||
| RA/RN/RT/RARS | 24 (20.5) | 48 (17.3) | 1.21 (0.61–2.38) |
| RCMD | 30 (25.6) | 74 (26.6) | 1.23 (0.67–2.72) |
| RAEB 1/2 | 34 (29.1) | 93 (33.5) | Ref |
| Therapy related | 14 (12.0) | 25 (9.0) | 1.30 (0.57–2.94) |
| MDS with del 5q | 10 (8.6) | 17 (6.1) | 1.05 (0.41–2.69) |
| MDS-U | 5 (4.3) | 21 (7.6) | 0.57 (0.19–1.72) |
| IPSS-R | |||
| Very low/Low | 52 (47.3) | 106 (41.4) | Ref |
| Intermediate | 18 (16.4) | 60 (23.4) | 0.58 (0.30–1.13) |
| High/Very high | 40 (36.4) | 90 (35.2) | 0.90 (0.52–1.57) |
| Missing | 29 |
aOR = adjusted odds ratio; BMI: body mass index; CI = confidence interval; HS: high school; MDS: myelodysplastic syndrome; NSAID: nonsteroidal anti-inflammatory drug.
ORs adjusted for age (continuous), sex, education (≤HS grad, some post HS, college grad), NSAID use, exposure to benzene, BMI 2 years ago (<25.0, 25.0–29.9, ≥30.0 kg/m2)
In the univariate Kaplan-Meier analysis, median overall survival times were not statistically significantly different in cases with (999 days (interquartile range: 435–2107 days)) or without (871 days (interquartile range: 392–1992 days) autoimmune diseases (p=0.91; Supplementary Figure). Likewise, the multivariable analysis did not reveal a statistically significant difference in overall survival (aHR=1.14, 95% CI: 0.64–2.03) after adjusting for age, sex, education, BMI, NSAID use and exposure to benzene. We did not have sufficient power to evaluate survival differences for individual autoimmune conditions.
Other chronic health conditions
Two of the other 11 adverse health conditions queried were significantly associated with MDS overall, including coronary heart disease (aOR=1.71, 95% CI: 1.27–2.31) and chronic fatigue syndrome (aOR=4.07, 95% CI: 1.51–11.0; Table 2). Similar results were observed across all subgroups examined. In the models evaluating latency (Supplementary Table 1), we found that the association with chronic fatigue syndrome was attenuated when we excluded cases diagnosed with the condition with 2 or 5 years prior to MDS diagnosis. All other associations remained similar in magnitude in the latency analysis. In the sex-stratified analysis, cirrhosis of the liver or liver damage (aOR=2.97, 95% CI: 1.03–8.58) was also associated with MDS in males, but not females (Supplementary Table 2).
Discussion
In this analysis, we have confirmed the association between MDS and autoimmune diseases in the aggregate and provided evidence that this association is strongest for hypothyroidism, with elevated but non-significant associations observed for inflammatory bowel disease, SLE, and polymyositis. Overall, we found that the co-occurrence of autoimmune disease and MDS arose in a much higher percentage of female cases than male cases. Associations were similar when we restricted our analysis to cases where the autoimmune disease was diagnosed at least 5 years before MDS, which suggests a temporal relationship where autoimmune disease precedes the development of MDS. Importantly, we did not observe a survival difference in cases with and without autoimmune disease. We also presented evidence for associations between cardiovascular disease and chronic fatigue syndrome and MDS.
The association we observed between history of any autoimmune disease and MDS is similar in magnitude to associations reported by Anderson et al. (n = 564 cases, OR=1.50, 95% CI 1.35–1.66) in an analysis of SEER-Medicare linked data [22] and by Wilson et al. (n = 849 cases, OR=1.50, 95% CI 1.10–2.0) in the General Practice Research Database.[24] An additional case-control study by Dalamaga et al. also noted an increased odds of autoimmune thyroiditis in cases compared with controls; no other autoimmune conditions were examined in that study.[29] Similar associations between AML and autoimmune disease overall and for specific autoimmune conditions have been observed.[22, 28, 30, 31] This consistency of association across studies, particularly within population-based studies, decreases the likelihood that the observed association is solely due to chance. It is possible that individuals with autoimmune disease may be more likely to be diagnosed with MDS due to increased interaction with the medical system. We did not have sufficient data available to rule out this possibility, although we would expect to see an association across all types of autoimmune disease if this was the primary explanation for the observed association.
While the overall association has been consistent across the few studies conducted to date, the specific autoimmune conditions associated with MDS have differed across studies. To date, associations have been reported for autoimmune thyroiditis,[25, 32, 33] rheumatoid arthritis,[22, 32] pernicious anemia,[22] psoriasis,[32] inflammatory bowel disease (IBD; including Crohn’s disease and/or ulcerative colitis),[24] and possibly SLE.[24] Of these, we observed a statistically significant association between MDS and hypothyroidism and non-significant, elevated odds ratios for inflammatory bowel disease and SLE. We did not observe an association between rheumatoid arthritis and MDS and did not capture information on pernicious anemia or psoriasis to permit examination of those conditions. Some of the inconsistency in the associations across studies is likely explained by the relatively small number of affected cases in each of the studies. Larger studies or pooled analyses will need to be conducted to more firmly establish causal relationships for individual autoimmune conditions.
The mechanism linking MDS and other hematologic malignancies with autoimmune conditions has not been firmly established; however, hypothesized explanations fall into three major categories, including shared genetic or environmental risk factors leading to both autoimmune disease and MDS, prior treatment for autoimmune diseases or damage to the bone marrow by the autoimmune disease causing the subsequent malignancy, or inflammatory mechanisms from the hematologic cancer resulting in the subsequent diagnosis of autoimmune disease.[22, 34] The evidence to support each of these hypotheses was recently reviewed in the context of CMML with some similar evidence for MDS.[34] With respect to the shared risk factor hypothesis, two firmly established risk factors for MDS, older age and male sex,[35] do not share similar associations with autoimmune disease. On the contrary, autoimmune disease is typically diagnosed at younger ages and is more common in females.[36] Similarly, while there may be some shared germline genetic susceptibility variants for autoimmune disease and MDS, these common variants are unlikely to serve as a sufficient oncogenic driver for the majority of cases. We did not capture information on treatment for autoimmune disease in study participants, so we were not able to evaluate associations based on prior treatment received, although a previous study reported increased risk of MDS and AML after treatment with azathioprine sodium for autoimmune disease.[37] While we did observe a stronger association between autoimmune disease overall and therapy-related MDS compared with de novo MDS (Table 3), several of the reported associations were strongly associated with MDS that was not classified as therapy-related. These findings suggest that treatment for autoimmune disease is also not likely to be the sole explanation for the observed associations.
The relative temporality of the two diagnoses is one epidemiologic clue that could be used to determine whether autoimmune disease is a cause or consequence of MDS. Our study and other epidemiologic studies have observed associations that persist after removing cases diagnosed within 5 years of diagnosis of malignancy, although some associations in a previous report were attenuated when recently diagnosed autoimmune conditions were removed.[22] The long latency period and preceding immune dysregulation associated with both MDS and autoimmune disease complicate the interpretation of this temporal analysis. For example, autoantibodies for rheumatoid arthritis and SLE may be observed years before diagnosis,[38, 39] while MDS is often preceded by clonal hematopoiesis of indeterminate potential (CHIP), defined as the presence of an expanded blood-cell clone in individuals without other hematologic abnormalities.[40] It is possible that chronic inflammatory processes associated with these somatic alterations may lead to both phenotypes.
Due to the advanced age at diagnosis for most MDS patients, co-morbidities are commonly reported and are associated with worse outcomes.[41–44] Few previous studies have evaluated the association between other chronic medical conditions and risk of MDS. While future studies will be required to confirm the observed association between cardiovascular disease and risk of MDS, a relationship between cardiovascular disease is biologically plausible given the shared link between inflammation, CHIP, and risk of both cardiovascular disease and MDS. In addition to being a strong risk factor for future hematologic malignancy, CHIP is associated with modest but significant associations with cardiovascular disease.[45] In a recent study of 200 patients undergoing hip arthroplasty, Hecker et al. confirmed associations between cardiovascular disease and clonal hematopoiesis and also reported a novel association between autoimmune disease and CHIP.[46] These findings lend support to a potential shared etiology for autoimmune disease, cardiovascular disease and MDS through inflammation and clonal hematopoiesis. We also observed an association between chronic fatigue syndrome and MDS. The attenuation of this association when we restricted to chronic fatigue syndrome diagnosed >2 or >5 years prior to MDS diagnosis increases the likelihood that the fatigue is caused by early stage MDS rather than being a causal association.
There are a number of strengths associated with our population-based study of MDS in Minnesota, including the rapid ascertainment of cases for a precipitously fatal disease to permit timely data collection. Cases were identified and approached shortly after diagnosis (median 140 days [range 7–1333 days]), and no proxy-interviews were conducted; rapid case ascertainment methods were utilized to minimize potential survival bias. Importantly, the use of the comprehensive Minnesota state driver’s license and identification card list following case identification allowed for controls to be selected from the at-risk adult population irrespective of exposure status.
Despite these strengths, there are also potential weaknesses that should be considered when interpreting our findings. Selection bias for both cases and controls is possible, given the response rates achieved in both case and control groups (59% and 49%, respectively). To partially address this concern, we compared limited demographic information across participants and non-participants, and our results suggested that participants were likely to be slightly older than non-participants, but no other meaningful differences were observed. Additionally, our analyses relied on self-reported data for medical conditions, thus it is important to consider potential recall bias. Using Behavioral Risk Factor Surveillance System data from Minnesota, we discovered similar prevalence estimates for cardiovascular disease (7% in ages 55–64 years; 16% in age 65+ years), arthritis (22% in adults), diabetes (12% in ages 55–64 years; 20% in age 65+ years), and hypertension (39% in ages 55–64 years; 56% in ages 65+ years) in that randomly selected, representative sample of Minnesota adults compared with the rates observed in controls in our study (Table 2).[47] Furthermore, unmeasured confounders could have contributed to the associations reported herein, although we have evaluated the potential confounding effects of established risk factors for MDS. The large majority of non-Hispanic white participants limits the generalizability of the data to other groups. Finally, the small number of cases with individual autoimmune conditions limits our statistical power and ability to draw definitive conclusions for specific associations.
In conclusion, we have validated the association between autoimmune disease overall and risk of MDS in this population-based study and provide additional evidence to suggest that this association is largely driven by hypothyroidism with potential elevated risk in individuals with inflammatory bowel disease, polymyositis and SLE. Further, we report an association between history of cardiovascular disease and risk of MDS which deserves further investigation due to the known associations between CHIP and both phenotypes.
Supplementary Material
Highlights.
MDS cases were significantly more likely to have a previous diagnosis of autoimmune disease compared with population controls.
Previous diagnosis with autoimmune disease did not impact overall survival following an MDS diagnosis.
Funding:
Supported by grants from the National Institutes of Health (R01 CA142714 to J.N.P.)
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest: The authors declare no potential conflicts of interest
Credit author statement
Amy Linabery: Conceptualization, Writing-Original draft preparation, Data analysis
Michelle Roesler: Data collection, Writing-Review and editing
Michaela Richardson: Data analysis, Writing-Review and editing
Erica Warlick: Data collection, Writing-Review and editing
Phuong Nguyen: Data collection, Writing-Review and editing
Adina Cioc: Data collection, Writing-Review and editing
Jenny Poynter: Conceptualization, Writing-Original draft preparation, Data analysis
References
- [1].Tefferi A, Vardiman JW. Myelodysplastic syndromes. N Engl J Med. 2009;361:1872–85. [DOI] [PubMed] [Google Scholar]
- [2].Volpe VO, Garcia-Manero G, Komrokji RS. Myelodysplastic Syndromes: A New Decade. Clin Lymphoma Myeloma Leuk. 2021. [DOI] [PubMed] [Google Scholar]
- [3].Rollison DE, Howlader N, Smith MT, Strom SS, Merritt WD, Ries LA, et al. Epidemiology of myelodysplastic syndromes and chronic myeloproliferative disorders in the United States, 2001–2004, using data from the NAACCR and SEER programs. Blood. 2008;112:45–52. [DOI] [PubMed] [Google Scholar]
- [4].Pedersen-Bjergaard J, Andersen MK, Christiansen DH. Therapy-related acute myeloid leukemia and myelodysplasia after high-dose chemotherapy and autologous stem cell transplantation. Blood. 2000;95:3273–9. [PubMed] [Google Scholar]
- [5].Boice JD Jr., Greene MH, Killen JY Jr., Ellenberg SS, Keehn RJ, McFadden E, et al. Leukemia and preleukemia after adjuvant treatment of gastrointestinal cancer with semustine (methyl-CCNU). N Engl J Med. 1983;309:1079–84. [DOI] [PubMed] [Google Scholar]
- [6].Pedersen-Bjergaard J, Daugaard G, Hansen SW, Philip P, Larsen SO, Rorth M. Increased risk of myelodysplasia and leukaemia after etoposide, cisplatin, and bleomycin for germ-cell tumours. Lancet. 1991;338:359–63. [DOI] [PubMed] [Google Scholar]
- [7].McLaughlin P, Estey E, Glassman A, Romaguera J, Samaniego F, Ayala A, et al. Myelodysplasia and acute myeloid leukemia following therapy for indolent lymphoma with fludarabine, mitoxantrone, and dexamethasone (FND) plus rituximab and interferon alpha. Blood. 2005;105:4573–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Surveillance, Epidemiology, and End Results (SEER) Program. SEER*Stat Database: Incidence - SEER 18 Regs Research Data, Nov 2016 Sub (2001–2014). Bethesda, Md: National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2017, based on the November 2016 submission. [Google Scholar]
- [9].Roman E, Smith A, Appleton S, Crouch S, Kelly R, Kinsey S, et al. Myeloid malignancies in the real-world: Occurrence, progression and survival in the UK’s population-based Haematological Malignancy Research Network 2004–15. Cancer Epidemiol. 2016;42:186–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Disperati P, Ichim CV, Tkachuk D, Chun K, Schuh AC, Wells RA. Progression of myelodysplasia to acute lymphoblastic leukaemia: implications for disease biology. Leuk Res. 2006;30:233–9. [DOI] [PubMed] [Google Scholar]
- [11].Castillo JJ, Reagan JL, Ingham RR, Furman M, Dalia S, Merhi B, et al. Obesity but not overweight increases the incidence and mortality of leukemia in adults: a meta-analysis of prospective cohort studies. Leuk Res. 2012;36:868–75. [DOI] [PubMed] [Google Scholar]
- [12].Descatha A, Jenabian A, Conso F, Ameille J. Occupational exposures and haematological malignancies: overview on human recent data. Cancer Causes Control. 2005;16:939–53. [DOI] [PubMed] [Google Scholar]
- [13].Kasim K, Levallois P, Abdous B, Auger P, Johnson KC. Lifestyle factors and the risk of adult leukemia in Canada. Cancer Causes Control. 2005;16:489–500. [DOI] [PubMed] [Google Scholar]
- [14].Poynter JN, Richardson M, Roesler M, Blair CK, Hirsch B, Nguyen P, et al. Chemical exposures and risk of acute myeloid leukemia and myelodysplastic syndromes in a population-based study. Int J Cancer. 2017;140:23–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Poynter JN, Richardson M, Blair CK, Roesler MA, Hirsch BA, Nguyen P, et al. Obesity over the life course and risk of acute myeloid leukemia and myelodysplastic syndromes. Cancer Epidemiol. 2015;40:134–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Lv L, Lin G, Gao X, Wu C, Dai J, Yang Y, et al. Case-control study of risk factors of myelodysplastic syndromes according to World Health Organization classification in a Chinese population. Am J Hematol. 2011;86:163–9. [DOI] [PubMed] [Google Scholar]
- [17].Travis LB, Li CY, Zhang ZN, Li DG, Yin SN, Chow WH, et al. Hematopoietic malignancies and related disorders among benzene-exposed workers in China. Leukemia & lymphoma. 1994;14:91–102. [DOI] [PubMed] [Google Scholar]
- [18].Pekmezovic T, Suvajdzic Vukovic N, Kisic D, Grgurevic A, Bogdanovic A, Gotic M, et al. A case-control study of myelodysplastic syndromes in Belgrade (Serbia Montenegro). Ann Hematol. 2006;85:514–9. [DOI] [PubMed] [Google Scholar]
- [19].Nisse C, Haguenoer JM, Grandbastien B, Preudhomme C, Fontaine B, Brillet JM, et al. Occupational and environmental risk factors of the myelodysplastic syndromes in the North of France. Br J Haematol. 2001;112:927–35. [DOI] [PubMed] [Google Scholar]
- [20].Strom SS, Gu Y, Gruschkus SK, Pierce SA, Estey EH. Risk factors of myelodysplastic syndromes: a case-control study. Leukemia. 2005;19:1912–8. [DOI] [PubMed] [Google Scholar]
- [21].Ma X, Lim U, Park Y, Mayne ST, Wang R, Hartge P, et al. Obesity, lifestyle factors, and risk of myelodysplastic syndromes in a large US cohort. Am J Epidemiol. 2009;169:1492–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Anderson LA, Pfeiffer RM, Landgren O, Gadalla S, Berndt SI, Engels EA. Risks of myeloid malignancies in patients with autoimmune conditions. Br J Cancer. 2009;100:822–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Kristinsson SY, Bjorkholm M, Hultcrantz M, Derolf AR, Landgren O, Goldin LR. Chronic immune stimulation might act as a trigger for the development of acute myeloid leukemia or myelodysplastic syndromes. J Clin Oncol. 2011;29:2897–903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Wilson AB, Neogi T, Prout M, Jick S. Relative risk of myelodysplastic syndromes in patients with autoimmune disorders in the General Practice Research Database. Cancer Epidemiol. 2014;38:544–9. [DOI] [PubMed] [Google Scholar]
- [25].Dalamaga M, Lekka A, Karmaniolas K, Stathopoulou E, Dionyssiou-Asteriou A. Is thyroid autoimmunity a risk factor for developing primary myelodysplastic syndrome? Cancer Causes Control. 2008;19:371–8. [DOI] [PubMed] [Google Scholar]
- [26].Anderson LA, Gadalla S, Morton LM, Landgren O, Pfeiffer R, Warren JL, et al. Population-based study of autoimmune conditions and the risk of specific lymphoid malignancies. Int J Cancer. 2009;125:398–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Vardiman JW, Thiele J, Arber DA, Brunning RD, Borowitz MJ, Porwit A, et al. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood. 2009;114:937–51. [DOI] [PubMed] [Google Scholar]
- [28].Johnson KJ, Blair CM, Fink JM, Cerhan JR, Roesler MA, Hirsch BA, et al. Medical conditions and risk of adult myeloid leukemia. Cancer Causes Control. 2012;23:1083–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Dalamaga M, Karmaniolas K, Papadavid E, Pelecanos N, Migdalis I. Association of thyroid disease and thyroid autoimmunity with multiple myeloma risk: a case-control study. Leukemia & lymphoma. 2008;49:1545–52. [DOI] [PubMed] [Google Scholar]
- [30].Ostgard LSG, Norgaard M, Pedersen L, Ostgard RD, Medeiros BC, Overgaard UM, et al. Autoimmune diseases, infections, use of antibiotics and the risk of acute myeloid leukaemia: a national population-based case-control study. Br J Haematol. 2018;181:205–14. [DOI] [PubMed] [Google Scholar]
- [31].Askling J, Brandt L, Lapidus A, Karlen P, Bjorkholm M, Lofberg R, et al. Risk of haematopoietic cancer in patients with inflammatory bowel disease. Gut. 2005;54:617–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Komrokji RS, Kulasekararaj A, Al Ali NH, Kordasti S, Bart-Smith E, Craig BM, et al. Autoimmune diseases and myelodysplastic syndromes. Am J Hematol. 2016;91:E280–3. [DOI] [PubMed] [Google Scholar]
- [33].Montoro J, Gallur L, Merchan B, Molero A, Roldan E, Martinez-Valle F, et al. Autoimmune disorders are common in myelodysplastic syndrome patients and confer an adverse impact on outcomes. Ann Hematol. 2018;97:1349–56. [DOI] [PubMed] [Google Scholar]
- [34].Ambinder AJ, Miller J, DeZern AE. Autoimmune disease in CMML-the chicken or the egg? Best practice & research Clinical haematology. 2020;33:101136. [DOI] [PubMed] [Google Scholar]
- [35].Ma X, Does M, Raza A, Mayne ST. Myelodysplastic syndromes: incidence and survival in the United States. Cancer. 2007;109:1536–42. [DOI] [PubMed] [Google Scholar]
- [36].Jacobson DL, Gange SJ, Rose NR, Graham NM. Epidemiology and estimated population burden of selected autoimmune diseases in the United States. Clin Immunol Immunopathol. 1997;84:223–43. [DOI] [PubMed] [Google Scholar]
- [37].Ertz-Archambault N, Kosiorek H, Taylor GE, Kelemen K, Dueck A, Castro J, et al. Association of Therapy for Autoimmune Disease With Myelodysplastic Syndromes and Acute Myeloid Leukemia. JAMA oncology. 2017;3:936–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].Arbuckle MR, McClain MT, Rubertone MV, Scofield RH, Dennis GJ, James JA, et al. Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N Engl J Med. 2003;349:1526–33. [DOI] [PubMed] [Google Scholar]
- [39].Kelmenson LB, Wagner BD, McNair BK, Frazer-Abel A, Demoruelle MK, Bergstedt DT, et al. Timing of Elevations of Autoantibody Isotypes Prior to Diagnosis of Rheumatoid Arthritis. Arthritis Rheumatol. 2020;72:251–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [40].Ebert BL, Libby P. Clonal Hematopoiesis Confers Predisposition to Both Cardiovascular Disease and Cancer: A Newly Recognized Link Between Two Major Killers. Ann Intern Med. 2018;169:116–7. [DOI] [PubMed] [Google Scholar]
- [41].Castelli R, Schiavon R, Deliliers GL. The impact of anaemia, transfusion dependency, comorbidities and polypharmacy in elderly patients with low-risk myelodysplastic syndromes. Med Oncol. 2018;35:33. [DOI] [PubMed] [Google Scholar]
- [42].Falantes JF, Marquez-Malaver FJ, Knight T, Calderon-Cabrera C, Martino ML, Gonzalez J, et al. The incorporation of comorbidities in the prognostication of patients with lower-risk myelodysplastic syndrome. Leukemia & lymphoma. 2017;58:1893–902. [DOI] [PubMed] [Google Scholar]
- [43].Balleari E, Salvetti C, Del Corso L, Filiberti R, Bacigalupo A, Bellodi A, et al. Age and comorbidities deeply impact on clinical outcome of patients with myelodysplastic syndromes. Leuk Res. 2015;39:846–52. [DOI] [PubMed] [Google Scholar]
- [44].Rozema J, Hoogendoorn M, Kibbelaar R, van den Berg E, Veeger N, van Roon E. Comorbidities and malignancies negatively affect survival in myelodysplastic syndromes: a population-based study. Blood Adv. 2021;5:1344–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [45].Jaiswal S, Fontanillas P, Flannick J, Manning A, Grauman PV, Mar BG, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371:2488–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [46].Hecker JS, Hartmann L, Riviere J, Buck MC, van der Garde M, Rothenberg-Thurley M, et al. CHIP & HIPs: Clonal Hematopoiesis is Common in Hip Arthroplasty Patients and Associates with Autoimmune Disease. Blood. 2021. [DOI] [PubMed] [Google Scholar]
- [47].(CDC) Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System Survey Data. In: U.S. Department of Health and Human Services Centers for Disease Control and Prevention, editor. Atlanta, Georgia: 2018. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.