Abstract
Introduction:
Children with congenital heart defects (CHD) have shorter life expectancy than the general population. Previous studies also suggest that patients with CHD have higher risk of cancer. This study aims to describe cancer-related mortality among patients with a history of CHD interventions using the Pediatric Cardiac Care Consortium (PCCC), a large US cohort of such patients.
Methods:
We performed a retrospective cohort study of individuals (<21 years) who underwent interventions for CHD in the PCCC from 1982 to 2003. Patients surviving their first intervention were linked to the National Death Index through 2020. Multivariable models assessed risk of cancer-related death, adjusting for age, sex, race, and ethnicity. Patients with/without genetic abnormalities (mostly Down syndrome [DS]) were considered separately, due to expected differential risk in cancer.
Results:
Among the 57,601 eligible patients in this study, cancer was the underlying or contributing cause of death for 208; with 20% among those with DS. Significantly increased risk of cancer-related death was apparent among patients with DS compared to the non-genetic group (aHR: 3.63, 95% confidence interval [CI]: 2.52–5.24, p < .001). For the group with non-genetic abnormalities, the highest association with cancer-related death compared to those with mild CHD was found among those with more severe CHD (severe two-ventricle aHR: 1.82, 95% CI: 1.04–3.20, p = .036, single-ventricle aHR: 4.68, 95% CI: 2.77–7.91, p < .001).
Conclusions:
Patients with more severe forms of CHD are at increased risk for cancer-related death. Despite our findings, we are unable to distinguish whether having CHD raises the risk of cancer or reduces survival.
Keywords: cancer epidemiology, congenital heart defects, long-term outcomes, mortality, pediatrics
1 |. INTRODUCTION
Congenital heart defects (CHD) impact approximately 1% of live births in the United States.1,2
Advances in treatment of these individuals have increased the number of adults living with history of operated CHD.3–5 Nevertheless, history of CHD remains a significant risk factor for post-intervention morbidities (including cancer) and premature death.6–9
Data from several international cohorts (Sweden, Canada, Denmark, and Taiwan) suggest an increased risk for cancer among patients with CHD by use of national registries. The relative risk for cancer occurrence among CHD patients has been reported to be consistently between 1.2- and 2.2-fold compared to matched population controls across all cancer types, after accounting for Down syndrome (DS).10–15 The reasons for such association between CHD and cancer are likely multifactorial, resulting from specific genetic predispositions16 and exposures during the course of CHD (medications,17,18 hypoxia,19 chronic stress,20 diagnostic studies, and surgical interventions that frequently include total or partial thymectomy21). One US study, using the California Birth Defects and Cancer Registry, has found similarly increased cancer risk among CHD patients, but yet, was not sufficient to examine the association between specific CHD conditions and cancer occurrence and mortality.22
In this study, we use the US-based Pediatric Cardiac Care Consortium (PCCC) registry to examine the association between risk of cancer-related death and CHD by severity and underlying pathophysiology. Given the large variation in genetic pathways, severity of symptoms, and exposures among patients with CHD, we hypothesize that there may be a differential risk for cancer by type and severity of CHD.
2 |. METHODS
2.1 |. Study population
This is a retrospective cohort study of patients from the PCCC who underwent interventions (catheterization and surgical) for CHD at less than 21 years of age, and subsequently died during follow-up. Initial PCCC interventions were conducted between 1982 and 2011. The PCCC contains clinical information from cardiac procedures performed at 47 US institutions on over 137,000 patients.23,24 PCCC centers discontinued data collection in 2011, and subsequent long-term survival status was acquired for patients enrolled up to April 15, 2003, date of implementation of the HIPAA (Health Insurance Portability and Accountability Act) rules that restricted the use of direct identifiers in the registry,24 either from the PCCC if the death event was reported to the registry or by linkage to the US National Death Index (NDI) up to the end of 2020.
2.2 |. Clinical information
Data collected from PCCC records included sex, CHD diagnosis, severity and physiologic classification, type of cardiac procedure, presence of a chromosomal or major genetic defect, and age at surgery (calculated). CHD severity classification in the PCCC categorizes conditions as mild, moderate, severe two-ventricle (2V) and single-ventricle (SV) type based on initial diagnosis and type of interventions as described before.6 Conditions that could not fit under any of these categories are listed as non-classifiable. This classification system was modified from the report of the Canadian Conference on the care of adults with CHD in 1996,25 and has been widely used by other investigators for the study of long-term outcomes of patients with CHD.26–29
For comparison analysis, we used the group of patients operated for patent ductus arteriosus (PDA) as reference, representing the least complex diagnosis in the cohort. This group includes patients with isolated PDA and without context of prematurity. Cancer types were divided into 12 categories (“Digestive/Gastrointestinal,” “Endocrine and Neuroendocrine,” “Eye,” “Genitourinary/Gynecologic/Breast,” “Head and Neck,” “Hematologic/Blood,” “Multiple Primary,” “Musculoskeletal/Skeletal,” “Central/Peripheral Nervous System,” “Respiratory/Thoracic,” “Skin,” “Unknown”) dictated by body location/organ systems according to descriptions provided by the National Cancer Institute (NCI).30 We further categorized each tumor as liquid or solid, hematologic or non-hematologic, and additional malignancies. Leukemia type (acute lymphoblastic leukemia [ALL], acute myeloid leukemia [AML], etc.) was individually assessed for those with a blood-related cancer. Results were stratified a priori into those for patients with and without genetic abnormalities.
2.3 |. Death ascertainment and US mortality data
Death events were obtained by linkage to the US NDI through December 31, 2020. Details about the linkage methodology have been previously described.6,31 Only patients with adequate identifiers were able to be linked to the NDI, and causes of death were obtained from the NDI Plus.32 Through this multiple iteration linkage, up to 39 years of potential follow-up was possible (1982–2020).
We queried those that could be matched to a death certificate for subjects with an NDI International Classification of Diseases—Version 9/10 (ICD-9/10) diagnosis of cancer as either underlying or contributing cause of death (COD). Additional malignancy identification was ascertained from contributing causes of death codes and required specific language stating, “Secondary malignancy…” or listing a malignancy in addition to their underlying COD. Additional cancer-specific information (cancer diagnosis type, hematologic) was categorized by expert review of NDI ICD-9/10 codes on death certificates by a co-author (Logan G. Spector). Death from any cause was considered in the calculation of follow-up time from CHD intervention. Incorporation of “cancer” as underlying or contributing COD allowed for capture of all deaths associated with a cancer diagnosis.21–23
2.4 |. Statistical analyses
Descriptive statistics were represented as counts and frequency (percentage) for categorical and medians with interquartile range (IQR) for continuous data. Continuous variables were assessed for normality via histograms, normal probability plots, and the Anderson–Darling test for normality showing right- or left-skewed distributions. Individuals were separated by genetic abnormality status in all analyses. To assess risk of cancer-related death (taken together as underlying and contributing cause) by demographic and CHD variables of interest, we constructed separate multivariable models for patients with and without genetic abnormalities. Specifically for genetic abnormalities, we showcased those with DS due to the high proportion of individuals and established links to cancer incidence.33 Models were adjusted for age at PCCC intervention, sex, race, and ethnicity. Adjusted hazard ratios (aHR) with 95% confidence intervals (CI) were presented accounting for age group at initial surgery, sex, race, ethnicity, and underlying CHD diagnosis, pathophysiology, and severity as defined in previous publications.34 Follow-up time was calculated from PCCC cardiac intervention to date of death or end of follow-up (December 31, 2020). Statistical comparisons were performed between groups with and without experience of a cancer-related death, and statistical significance was assessed at the .05 level. All analyses were performed using SAS version 9.4 (Cary, NC).
3 |. RESULTS
3.1 |. Study population
The study cohort included 57,601 PCCC patients with an intervention for CHD between 1982 and 2011. Of these, 7532 (13%) had a genetic abnormality, and more specifically 4949 had DS (Figure 1). Among the 50,069 patients without a known genetic abnormality, we observed 7500 deaths, with 152 of them as being cancer-related. Among the 877 DS patients dying by the end of 2020, 42 were cancer-related.
FIGURE 1.
Flow diagram of study cohort, by genetic abnormalities and cancer-related death status.
3.2 |. Patient characteristics
Patient characteristics are reported in Table 1 by genetic abnormality status. Of those with a genetic abnormality, the majority had DS (65%), followed by DiGeorge syndrome (9%). Individuals with DS were more likely to be female, have a CHD intervention in their first year of life (70%), and have a left-to-right shunt as CHD primary diagnosis. Among DS patients with a cancer-related death, 92% were diagnosed with “hematologic/blood” cancers in contrast to all other NCI designations displayed in Table 2. Median age at cancer death was observed at 12.7 years (IQR: 7.5–20.9), nearly 7 years earlier than those with cancer-related death, but without a genetic abnormality.
TABLE 1.
Characteristics of congenital heart intervention patients in study cohort, by genetic abnormality status.
| Genetic abnormalities (N = 7532) |
||||
|---|---|---|---|---|
| Variable | No genetic abnormalities (N = 50,069) |
Down syndrome n = 4949 |
Othera n = 2583 |
p-Valueb |
| Sex | <.001 | |||
| Female | 23,157 (46.3) | 2574 (52.0) | 1333 (51.6) | |
| Male | 26,912 (53.7) | 2375 (48.0) | 1250 (48.4) | |
| Race | <.001 | |||
| White | 14,693 (29.4) | 1537 (31.1) | 788 (30.5) | |
| Black | 2883 (5.8) | 343 (6.9) | 170 (6.6) | |
| Otherc | 521 (1.0) | 47 (0.9) | 32 (1.2) | |
| Unknown | 31,972 (63.9) | 3022 (61.1) | 1593 (61.7) | |
| Age at first intervention | <.001 | |||
| Median (IQR), years | 1.10 (0.10–5.15) | 0.51 (0.31–1.38) | 0.45 (0.01–3.83) | |
| Neonate (0 to <28 days) | 12,078 (24.1) | 321 (6.5) | 860 (33.3) | |
| Infant (28 days to <1 year) | 12,271 (24.5) | 3141 (63.5) | 696 (27.0) | |
| Young child (1 to <10 years) | 18,856 (37.7) | 1220 (24.7) | 732 (28.3) | |
| Older child | 6864 (13.7) | 267 (5.4) | 295 (11.4) | |
| Primary diagnosis | <.001 | |||
| PDA | 4583 (9.2) | 294 (5.9) | 104 (4.0) | |
| Left to right shuntd | 11,200 (22.4) | 3968 (80.2) | 398 (15.4) | |
| LHOL | 7765 (15.5) | 69 (1.4) | 615 (23.8) | |
| APVR | 1810 (3.6) | 7 (0.1) | 38 (1.5) | |
| RVOTO | 7325 (14.6) | 280 (5.7) | 464 (18.0) | |
| TGA physiology | 3423 (6.8) | 7 (0.1) | 29 (1.1) | |
| Truncus arteriosus | 556 (1.1) | 5 (0.1) | 141 (5.5) | |
| Miscellaneous physiologye | 7332 (14.6) | 231 (4.7) | 621 (24.0) | |
| Single ventricle | 6075 (12.1) | 88 (1.8) | 173 (6.7) | |
| Severity | <.001 | |||
| Mild | 13,988 (27.9) | 1667 (33.7) | 428 (16.6) | |
| Moderate | 16,396 (32.8) | 3036 (61.4) | 994 (38.5) | |
| Severe 2V | 8522 (17.0) | 111 (2.2) | 699 (27.1) | |
| Single ventricle | 6075 (12.1) | 88 (1.8) | 173 (6.7) | |
| Unclassifiable | 5088 (10.2) | 47 (1.0) | 289 (11.2) | |
| Cyanotic | <.001 | |||
| No | 29,053 (58.0) | 4494 (90.8) | 1283 (49.7) | |
| Yes | 17,483 (34.9) | 427 (8.6) | 909 (35.2) | |
| Unclassifiable | 3533 (7.1) | 28 (0.6) | 391 (15.1) | |
Abbreviations: APVR, anomalous pulmonary venous return; LHOL, left heart obstructive lesion; PDA, patent ductus arteriosus; RVOTO, right ventricular outflow tract obstruction; TGA, transposition of the great arteries.
Includes Marfan, DiGeorge, Noonan, Williams, Turners, and miscellaneous other genetic syndromes and conditions.
p-Values compare overall group with and without genetic abnormalities.
Includes “Asian,” “Native American,” “more than one race,” “Pacific Islander,” and “Other.”
Other than PDA.
Miscellaneous lesions not classifiable by a single major pathophysiology.
TABLE 2.
Characteristics of cancer-related deaths in the cohort by the presence of genetic abnormalities.
| Genetic abnormalities (N = 7532) |
||||
|---|---|---|---|---|
| Variable | No genetic abnormalities (N = 50,069) |
Down syndrome n = 4949 |
Othera n = 2583 |
p-Valueb |
| Cancer-related deaths | 152 | 42 | 14 | |
| Sex—female | 64 (42.0) | 20 (47.6) | 8 (57.1) | .309 |
| Race | .013 | |||
| White | 52 (34.2) | 12 (28.6) | 2 (14.3) | |
| Black | 10 (6.6) | – | 2 (14.3) | |
| Otherc | 3 (2.0) | 2 (4.8) | – | |
| Unknown | 87 (57.2) | 28 (66.7) | 10 (71.4) | |
| Median age at first intervention, years | 1.31 (0.03–6.27) | 0.63 (0.25–2.34) | 1.91 (0.17–9.67) | .635 |
| Median age at death, years (IQR) | 19.8 (9.8–27.1) | 12.7 (7.5–20.9) | 21.7 (8.7–28.8) | .060 |
| Broad cancer diagnosis type | <.001 | |||
| Solid | 99 (65.1) | 9 (21.4) | 10 (71.4) | |
| Liquid | 37 (24.4) | 32 (76.2) | 3 (21.4) | |
| Unclassifiable | 16 (10.5) | 1 (2.4) | 1 (7.2) | |
| Cancer organ/location (NCI designations) | .002 | |||
| Digestive/gastrointestinal | 31 (20.8) | 1 (2.4) | 1 (7.2) | |
| Endocrine and neuroendocrine | 2 (1.3) | – | 1 (7.2) | |
| Eye | 1 (0.7) | – | – | |
| Genitourinary/gynecologic/breast | 9 (6.0) | – | 1 (7.2) | |
| Head and neck | 2 (1.3) | 1 (1.8) | – | |
| Hematologic/blood | 58 (37.7) | 39 (92.9) | 6 (42.9) | |
| Multiple primary | 4 (2.6) | – | – | |
| Musculoskeletal/skeletal | 7 (4.6) | – | 2 (14.3) | |
| Central/peripheral nervous system | 20 (13.0) | – | 2 (14.3) | |
| Respiratory/thoracic | 4 (2.6) | – | – | |
| Skin | 3 (2.0) | – | – | |
| Unknown | 11 (7.8) | 1 (2.4) | 1 (7.2) | |
| Additional malignancyd | .742 | |||
| Yes | 13 (8.6) | 2 (4.8) | 2 (14.3) | |
| CHD primarydiagnosis | <.001 | |||
| PDA | 14 (9.2) | 3 (7.1) | – | |
| Left to right shunte | 19 (12.5) | 31 (73.8) | 2 (14.3) | |
| LHOL | 21 (13.8) | – | 7 (50.0) | |
| APVR | 6 (4.0) | – | – | |
| RVOTO | 14 (9.2) | 4 (9.5) | 2 (14.3) | |
| TGA physiology | 8 (5.3) | – | – | |
| Truncus arteriosus | 1 (0.7) | – | – | |
| Miscellaneous physiologyf | 30 (19.7) | 3 (7.1) | 2 (14.3) | |
| Single ventricle | 39 (25.7) | 1 (2.4) | – | |
| CHD severity | <.001 | |||
| Mild | 28 (18.4) | 8 (19.1) | 2 (14.3) | |
| Moderate | 34 (22.4) | 29 (69.1) | 6 (42.9) | |
| Severe 2V | 29 (19.4) | 3 (7.1) | 3 (21.4) | |
| Single ventricle | 39 (25.7) | 1 (2.4) | – | |
| Unclassifiable | 22 (14.5) | 1 (2.4) | 3 (21.4) | |
| Cyanotic | .003 | |||
| No | 74 (48.7) | 34 (80.9) | 8 (57.1) | |
| Yes | 66 (43.4) | 8 (19.1) | 3 (21.4) | |
| Unclassifiable | 12 (7.9) | 0 (0.0) | 3 (21.4) | |
Abbreviations: APVR, anomalous pulmonary venous return; CHD, congenital heart defect; LHOL, left heart obstructive lesion; NCI, National Cancer Institute; PDA, patent ductus arteriosus; RVOTO, right ventricular outflow tract obstruction; TGA, transposition of the great arteries.
Includes Marfan, DiGeorge, Noonan, Williams, Turner, and miscellaneous other genetic syndromes and conditions.
p-Values compare overall group with and without genetic abnormalities.
Includes “Asian,” “Native American,” “More than one,” “Pacific Islander,” and “Other.”
Conditions indicated as “secondary malignant neoplasms” on their death record or listed multiple malignancies as contributing causes of death.
Other than PDA.
Miscellaneous lesions not classifiable by a single major pathophysiology.
Patients without a genetic abnormality were operated at older ages (>1 year of age at CHD intervention) and had more severe CHD diagnoses (severe 2V, single ventricle) than DS patients. Among the cancer-related deaths in this group, almost two-thirds had a solid tumor diagnosis on their death certificate, and 43.4% had a cyanotic CHD (Table 2). Interestingly, the second most common cancer diagnosis in this group was “digestive/gastrointestinal,” mainly attributed to primary hepatic malignancies (n = 17). These included, liver cell carcinoma (n = 8), intrahepatic bile duct carcinoma (n = 3), hepatoblastoma (n = 3), and non-otherwise specified hepatic malignancies (n = 3). Among the patients with primary hepatic malignancy other than hepatoblastoma, eight (57%) had history of Fontan procedure, with all of them dying from liver cell carcinoma (n = 5) or the related intrahepatic bile duct carcinoma (n = 2), and one from a hepatic malignancy non-otherwise specified.
In both groups, the “hematologic/blood” cancer category was primarily comprised of leukemia diagnoses (“AML,” “ALL,” and “not otherwise specified” [NOS]). Additional malignancies to the main cancer-related death diagnosis for each group were rare, and the order of their occurrence in relationship to the leukemia diagnosis was not available.
3.3 |. Multivariable analysis
For all patients, the risk of dying from cancer rose with increasing severity of CHD when compared to the mild CHD (Table 3). Using as reference, the group with patent ductus arteriosus (PDA), those with single ventricle had the highest rate of cancer-related death, and conferred a 174% higher risk (aHR: 2.74, 95% CI: 1.43–5.24, p = .002) among the non-genetic patients. Among the individual cancer diagnoses, stood out the association between Fontan physiology and liver cancer. More specifically, patients with Fontan physiology (n = 3096) had a relative risk (RR) of 14.18 (95% CI: 4.98–40.41) to succumb from liver malignancy (other than hepatoblastoma or metastatic disease) when compared with all other CHD patients combined (except those with PDA) and 11.64 (95% CI: 1.43–94.57) when compared with patients with PDA closure, which we use as the control group closest to the general population. However, patients with non-Fontan CHD (except PDA) had a RR of 0.92 (95% CI: 0.11–7.55) compared to the PDA group. Cyanotic CHD (cyanotic vs. non-cyanotic aHR: 1.62, 95% CI: 1.13–2.32, p = .009) at first intervention was also an independent risk factor associated with cancer-related death in the non-genetic group. Infantile (aHR: 0.61, 95% CI: 0.38–0.97, p = .037) and young child age groups (aHR: 0.61, 95% CI: 0.40–0.92, p = .018) at CHD surgery were associated with decreased risk for cancer-related death compared to those required surgery as neonates, but this trend did not persist in adolescence. No significant associations were seen between sex, race, or ethnicity and cancer-related deaths.
TABLE 3.
Risk of cancer death in the PCCC cohort among patients without genetic abnormalities and those with a Down syndrome diagnosis (N = 55,018).
| Non-genetic | Down syndrome | |||||||
|---|---|---|---|---|---|---|---|---|
| Risk factor | At risk N = 50,069 |
Cancer deaths N = 152 |
aHR | 95% CI | At risk N = 4949 |
Cancer deaths N = 42 |
aHR | 95% CI |
| Overall | 1.0 | – | 3.63 | 2.52–5.24 | ||||
| Demographic | ||||||||
| Sex | ||||||||
| Female | 23,157 | 64 | 1.0 | – | 2574 | 20 | 1.0 | – |
| Male | 26,912 | 88 | 1.15 | 0.83–1.59 | 2375 | 22 | 1.17 | 0.64–2.15 |
| Race | ||||||||
| White | 14,693 | 52 | 1.0 | – | 1537 | 12 | 1.0 | – |
| Black | 2883 | 10 | 1.05 | 0.53–2.07 | 343 | 0 | N/A | – |
| Othera | 521 | 3 | N/A | – | 47 | 2 | N/A | – |
| Unknown | 31,972 | 87 | 0.68 | 0.44–1.04 | 3022 | 28 | 1.05 | 0.46–2.42 |
| Ethnicity | ||||||||
| Not Hispanic | 15,118 | 49 | 1.0 | – | 1627 | 11 | 1.0 | – |
| Hispanic | 1358 | 3 | N/A | – | 169 | 7 | 5.43 | 1.87–15.77 |
| Unknown | 33,593 | 100 | 1.32 | 0.86–2.05 | 3153 | 24 | 1.03 | 0.43–2.46 |
| Clinical | ||||||||
| Primary diagnosis | ||||||||
| PDA | 4583 | 14 | Ref | – | 294 | 3 | Ref | – |
| Left to right shuntb | 11,200 | 19 | 0.53 | 0.26–1.05 | 3968 | 31 | 1.02 | 0.30–3.44 |
| LHOL | 7765 | 21 | 0.74 | 0.37–1.49 | 69 | 0 | N/A | – |
| APVR | 1810 | 6 | 1.03 | 0.39–2.72 | 7 | 0 | N/A | – |
| RVOTO | 7325 | 14 | 0.58 | 0.27–1.23 | 280 | 4 | 1.38 | 0.30–6.42 |
| TGA physiology | 3423 | 8 | 0.69 | 0.28–1.73 | 7 | 0 | N/A | – |
| Truncus arteriosus | 556 | 1 | N/A | 0.09–5.52 | 5 | 0 | N/A | – |
| Miscellaneousc | 7332 | 30 | 1.25 | 0.65–2.39 | 231 | 3 | 1.09 | 0.21–5.66 |
| Single ventricle | 6075 | 39 | 2.74 | 1.43–5.24 | 88 | 1 | 1.45 | 0.14–15.38 |
| CHD severity | ||||||||
| Mild | 13,988 | 28 | Ref | – | 1667 | 8 | Ref | – |
| Moderate | 16,396 | 34 | 1.01 | 0.61–1.67 | 3036 | 29 | 2.52 | 1.13–5.58 |
| Severe 2V | 8522 | 29 | 1.82 | 1.04–3.20 | 111 | 3 | N/A | – |
| Single ventricle | 6075 | 39 | 4.68 | 2.77–7.91 | 88 | 1 | N/A | – |
| Unclassifiable | 5088 | 22 | 2.01 | 1.13–3.58 | 47 | 1 | 5.41 | 0.65–44.98 |
| Cyanotic | ||||||||
| No | 29,053 | 74 | Ref | – | 4494 | 34 | Ref | – |
| Yes | 17,483 | 66 | 1.62 | 1.13–2.32 | 427 | 8 | 1.94 | 0.84–4.50 |
| Unclassifiable | 3533 | 12 | 1.37 | 0.74–2.53 | 28 | 0 | N/A | – |
| Age at first intervention | ||||||||
| Neonate (0 to <28 days) | 12,078 | 41 | Ref | – | 321 | 7 | Ref | – |
| Infant (28 days to <1 year) | 12,271 | 30 | 0.61 | 0.38–0.97 | 3141 | 17 | 0.19 | 0.08–0.47 |
| Young child (1 to <10 years) | 18,856 | 51 | 0.61 | 0.40–0.92 | 1220 | 15 | 0.39 | 0.16–0.95 |
| Older child (10 to <21 years) | 6864 | 30 | 1.05 | 0.65–1.69 | 267 | 3 | N/A | – |
Note: N/A = not available because of wide confidence intervals and interpretability issues when <5 deaths contributed to aHRs. Numbers in bold indicate statistical significance at p < .05.
Abbreviations: aHR, adjusted hazard ratio; APVR, anomalous pulmonary venous return; CHD, congenital heart defect; CI, confidence interval; LHOL, left heart obstructive lesion; PCCC, Pediatric Cardiac Care Consortium; PDA, patent ductus arteriosus; RVOTO, right ventricular outflow tract obstruction; TGA, transposition of the great arteries.
Includes “Asian,” “Native American,” “More than one,” “Pacific Islander,” and “Other.”
Other than PDA.
Miscellaneous lesions not classifiable by a single major pathophysiology.
Due to the wide range of clinical conditions among those with genetic abnormalities and the very unique risks for the largest group among them with DS, we focused our analysis only on this group. Overall, an increased risk of cancer-related death was apparent among patients with DS (aHR: 3.63, 95% CI: 2.52–5.24, p < .001). Similar trends among these patients were noted as in the group without genetic abnormalities; however, due to the small number of patients and cancer-related events, we could not assess the risk for patients with some CHD physiology and severity categories. Interestingly for this subgroup, there was a five-fold higher risk of cancer-related death among those reported to be of Hispanic ethnicity than those who were not (aHR: 5.43, 95% CI: 1.87–14.05, p = .002); however, this part of the analysis was limited by the lack of race/ethnicity information for the majority of the cohort.
4 |. DISCUSSION
Previous work in the PCCC has indicated that there is higher cancer-related mortality in those with CHD compared to the general population.34 Our current study shows that the risk of cancer-related death in those with a history of interventions for CHD is higher in patients with more severe and cyanotic types of CHD, as well with DS. Further, Hispanic descent was indicative of higher cancer-related mortality in those with DS.
Recent work on the relationship between CHD and cancer incidence in Swedish cohorts found that subjects with CHD diagnosis had more than double the risk of cancer,10,12,13 with the highest cancer incidence among those operated in infancy (likely including mostly subjects with severe and cyanotic forms of CHD).13 Despite increased incidence for cancer in these children (<17 years), earlier pediatric diagnoses have been broadly linked with higher rates of survival compared to diagnoses at older ages,35 with consideration of type of cancer. In our study, we observed a protective effect against eventual cancer-related mortality for children operated on for CHD during infancy and early childhood ages. These somewhat conflicting findings could suggest that although incidence of cancer may be higher post-CHD intervention in the neonatal age, subsequent mortality due to cancer remains relatively low. Other studies have linked childhood cancer as a strong risk factor for second malignancies later in life.36–41 Despite this, most research observes stratification of this increased risk by initial cancer treatment modalities (radiation therapy, chemotherapy, etc.), genetic makeup, and era of previous cancer diagnosis (improvements in treatment over time) allowing for variation in outcomes. Furthermore, second or later in life cancer diagnoses do not implicate absolute death from that cause and may be absent from death records analyzed in our study.
Findings from California and Swedish cohorts showcased a substantially higher incidence for cancer with more severe forms of CHD.12,22 We observed this similar association within our cohort for cancer mortality, in both non-genetic and DS patients. The heightened risk in single ventricle and more severe forms of CHD could be explained by worsened prognosis when cancer diagnosis is superimposed to severe CHD. Alternatively, severe forms of CHD may be specifically linked to risk for certain malignancies such as the case of single ventricle with Fontan palliation and liver cancer. The increased incidence of death from liver cancer among patients with Fontan physiology in our cohort is consistent with previous reports42–44; however, our current analysis cannot fully address these questions without specific information on cancer diagnosis and timing.
Premature death (by any cause) has been frequently observed in the literature as more common in those with genetic abnormalities.6,34,45–48 Specifically, death from cancer is more probable in those with DS due to their increased risk for leukemia compared to the average person.49 In some cases, cancer incidence has been observed as two- to three-fold as common in those with CHD and up to 10–20-fold for those with DS.50 Our finding of increased risk of cancer-related death in those with CHD and genetic abnormalities is congruent with a previous PCCC finding34 and aligned with cancer incident risk studies, especially in our cohort of primarily leukemic cases.
The PCCC is the oldest clinical registry for CHD diagnoses and procedures in the United States. This allows us the possibility to examine the association of CHD diagnosis and treatment on cancer-related mortality on a long-term scale. Implications for research in this cohort are widespread and allow for targeted surveillance of factors in CHD care and treatment over time that can potentially affect mortality later in life. The NDI remains the most essential source of data for characterizing mortality35 by using the listed underlying and contributing causes of death. The matching of the NDI Plus records with the PCCC)21,22 allowed for up to 30 years of follow-up in these patients. No other registry confers this advantage for looking at the long-term outcomes of CHD treatment modalities by types of CHD.
Our study is not without limitations. Inherent to a retrospective, registry-based study, we are only able to examine variables collected at the time of CHD intervention at PCCC-participating hospitals. Most notably, additional procedures outside of the registry, use of medications, environmental, socioeconomic, and lifestyle exposures are not available within this dataset. Another limitation of our study is the lack of detailed genetic diagnoses, as it relates to single gene abnormalities. Additionally, our study lacks information on the timing and details of cancer diagnosis and treatment. In studies of etiology, it is preferred to know incidence rather than mortality to assess the association of CHD diagnosis and treatment with the cancer diagnosis. Although incident cancer is unknown, risk of cancer-related death and factors increasing/decreasing this outcome can be estimated. Without linkage to a cancer registry database, treatment decisions and acceptability are unknown in this cohort. Precision of COD from the NDI Plus records is also another potential drawback to our study. Although detailed death records such as this have been utilized effectively in past studies of cancer death, human error through subjective ICD-9/10 coding decisions cannot be ruled out. However, additional review of death codes by a cancer epidemiologist, as well as validation of using death certificate data in other cohorts34 (including in cancer deaths), gives us confidence that deaths were correctly categorized for our population and within the known limitations of death certificates.
Our study demonstrates increased risk for cancer-related death among patients with more severe forms of CHD. Additional studies linking data from nationally representative cancer registries with CHD datasets could help to elucidate the incidence and timing of cancer occurrence in relationship to diagnosis and surgical treatment of CHD.
ACKNOWLEDGMENTS
We thank the program directors and data collection coordinators from the participating PCCC centers; without their effort and dedication, this work could not have been completed. This study was supported by the National Heart, Lung, and Blood Institute R01 (HL122392) and the Department of Defense (PR180683). Research reported in this publication was also supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) under Award Number T32HD095134. This project also benefited from support provided by the Minnesota Population Center (P2CHD041023), which also receives funding from NICHD.
Abbreviations:
- ALL
acute lymphoblastic leukemia
- AML
acute myeloid leukemia
- CHD
congenital heart defect
- COD
cause of death
- DS
Down syndrome
- ICD-9/10
International Classification of Diseases—Version 9/10
- NCI
National Cancer Institute
- NDI
National Death Index
- PCCC
Pediatric Cardiac Care Consortium
Footnotes
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
REFERENCES
- 1.Hoffman JIE, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol. 2002;39(12):1890–1900. doi: 10.1016/s0735-1097(02)01886-7 [DOI] [PubMed] [Google Scholar]
- 2.Reller MD, Strickland MJ, Riehle-Colarusso T, Mahle WT, Correa A. Prevalence of congenital heart defects in metropolitan Atlanta, 1998–2005. J Pediatr. 2008;153(6):807–813. doi: 10.1016/j.jpeds.2008.05.059 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mandalenakis Z, Rosengren A, Skoglund K, Lappas G, Eriksson P, Dellborg M. Survivorship in children and young adults with congenital heart disease in Sweden. JAMA Intern Med. 2017;177(2):224–230. doi: 10.1001/jamainternmed.2016.7765 [DOI] [PubMed] [Google Scholar]
- 4.Diller GP, Breithardt G, Baumgartner H. Congenital heart defects in adulthood. Dtsch Arzteblatt Int. 2011;108(26):452–459. doi: 10.3238/arztebl.2011.0452 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Scott M, Neal AE. Congenital heart disease. Prim Care. 2021;48(3):351–366. doi: 10.1016/j.pop.2021.04.005 [DOI] [PubMed] [Google Scholar]
- 6.Spector LG, Menk JS, Knight JH, et al. Trends in long-term mortality after congenital heart surgery. J Am Coll Cardiol. 2018;71(21):2434–2446. doi: 10.1016/j.jacc.2018.03.491 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Oster ME. Comorbidities among young adults with congenital heart defects: results from the congenital heart survey to recognize outcomes, needs, and well-being—Arizona, Arkansas, and Metropolitan Atlanta, 2016–2019. MMWR Morb Mortal Wkly Rep. 2021;70:197–201. doi: 10.15585/mmwr.mm7006a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mahle WT, Wernovsky G. Long-term developmental outcome of children with complex congenital heart disease. Clin Perinatol. 2001;28(1):235–247. doi: 10.1016/s0095-5108(05)70077-4 [DOI] [PubMed] [Google Scholar]
- 9.Razzaghi H, Oster M, Reefhuis J. Long-term outcomes in children with congenital heart disease: National Health Interview Survey. J Pediatr. 2015;166(1):119–124. doi: 10.1016/j.jpeds.2014.09.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mandalenakis Z, Karazisi C, Skoglund K, et al. Risk of cancer among children and young adults with congenital heart disease compared with healthy controls. JAMA Netw Open. 2019;2(7):e196762. doi: 10.1001/jamanetworkopen.2019.6762 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gurvitz M, Ionescu-Ittu R, Guo L, et al. Prevalence of cancer in adults with congenital heart disease compared with the general population. Am J Cardiol. 2016;118(11):1742–1750. doi: 10.1016/j.amjcard.2016.08.057 [DOI] [PubMed] [Google Scholar]
- 12.Kampitsi CE, Mogensen H, Feychting M, Tettamanti G. The relationship between congenital heart disease and cancer in Swedish children: a population-based cohort study. PLoS Med. 2022;19(2):e1003903. doi: 10.1371/journal.pmed.1003903 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Karazisi C, Dellborg M, Mellgren K, et al. Risk of cancer in young and older patients with congenital heart disease and the excess risk of cancer by syndromes, organ transplantation and cardiac surgery: Swedish Health Registry Study (1930–2017). Lancet Reg Health Eur. 2022;18:100407. doi: 10.1016/j.lanepe.2022.100407 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lee YS, Chen YT, Jeng MJ, et al. The risk of cancer in patients with congenital heart disease: a nationwide population-based cohort study in Taiwan. PLoS One. 2015;10(2):e0116844. doi: 10.1371/journal.pone.0116844 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Olsen M, Garne E, Sværke C, et al. Cancer risk among patients with congenital heart defects: a nationwide follow-up study. Cardiol Young. 2014;24(1):40–46. doi: 10.1017/S1047951112002144 [DOI] [PubMed] [Google Scholar]
- 16.Morton SU, Shimamura A, Newburger PE, et al. Association of damaging variants in genes with increased cancer risk among patients with congenital heart disease. JAMA Cardiol. 2021;6(4):457–462. doi: 10.1001/jamacardio.2020.4947 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Pottegård A, Pedersen SA, Schmidt SAJ, Hölmich LR, Friis S, Gaist D. Association of hydrochlorothiazide use and risk of malignant melanoma. JAMA Intern Med. 2018;178(8):1120–1122. doi: 10.1001/jamainternmed.2018.1652 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hicks BM, Filion KB, Yin H, Sakr L, Udell JA, Azoulay L. Angiotensin converting enzyme inhibitors and risk of lung cancer: population based cohort study. BMJ. 2018;363:k4209. doi: 10.1136/bmj.k4209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Muz B, de la Puente P, Azab F, Azab AK. The role of hypoxia in cancer progression, angiogenesis, metastasis, and resistance to therapy. Hypoxia. 2015;3:83. doi: 10.2147/HP.S93413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dai S, Mo Y, Wang Y, et al. Chronic stress promotes cancer development. Front Oncol. 2020;10:1492. doi: 10.3389/fonc.2020.01492 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gudmundsdottir J, Söderling J, Berggren H, et al. Long-term clinical effects of early thymectomy: associations with autoimmune diseases, cancer, infections, and atopic diseases. J Allergy Clin Immunol. 2018;141(6):2294–2297. doi: 10.1016/j.jaci.2018.01.037 [DOI] [PubMed] [Google Scholar]
- 22.Collins RT, Von Behren J, Yang W, et al. Congenital heart disease complexity and childhood cancer risk. Birth Defects Res. 2018;110(17):1314–1321. doi: 10.1002/bdr2.1390 [DOI] [PubMed] [Google Scholar]
- 23.Vinocur JM, Moller JH, Kochilas LK. Putting the Pediatric Cardiac Care Consortium in context: evaluation of scope and case mix compared with other reported surgical datasets. Circ Cardiovasc Qual Outcomes. 2012;5(4):577–579. doi: 10.1161/CIRCOUTCOMES.111.964841 [DOI] [PubMed] [Google Scholar]
- 24.Moller JH, Hills CB, Pyles LA. A multi-center cardiac registry. A method to assess outcome of catheterization intervention or surgery. Prog Pediatr Cardiol. 2005;20(1):7–12. doi: 10.1016/j.ppedcard.2004.12.009 [DOI] [Google Scholar]
- 25.Connelly MS, Webb GD, Somerville J, et al. Canadian Consensus Conference on Adult Congenital Heart Disease 1996. Can J Cardiol. 1998;14(3):395–452. [PubMed] [Google Scholar]
- 26.Warnes CA, Liberthson R, Danielson GK, et al. Task force 1: the changing profile of congenital heart disease in adult life. J Am Coll Cardiol. 2001;37(5):1170–1175. doi: 10.1016/s0735-1097(01)01272-4 [DOI] [PubMed] [Google Scholar]
- 27.Nieminen HP, Jokinen EV, Sairanen HI. Late results of pediatric cardiac surgery in Finland: a population-based study with 96% follow-up. Circulation. 2001;104(5):570–575. doi: 10.1161/hc3101.093968 [DOI] [PubMed] [Google Scholar]
- 28.Erikssen G, Liestøl K, Seem E, et al. Achievements in congenital heart defect surgery: a prospective, 40-year study of 7038 patients. Circulation. 2015;131(4):337–346. doi: 10.1161/CIRCULATIONAHA.114.012033 [DOI] [PubMed] [Google Scholar]
- 29.Raissadati A, Nieminen H, Jokinen E, Sairanen H. Progress in late results among pediatric cardiac surgery patients: a population-based 6-decade study with 98% follow-up. Circulation. 2015;131(4):347–353. doi: 10.1161/CIRCULATIONAHA.114.011190 [DOI] [PubMed] [Google Scholar]
- 30.Cancers by body location/system—NCI. NCI; Published January 1, 1980. Accessed December 9, 2023. https://www.cancer.gov/types/by-body-location [Google Scholar]
- 31.Spector LG, Menk JS, Vinocur JM, et al. In-hospital vital status and heart transplants after intervention for congenital heart disease in the Pediatric Cardiac Care Consortium: completeness of ascertainment using the National Death Index and United Network for Organ Sharing Datasets. J Am Heart Assoc Cardiovasc Cerebrovasc Dis. 2016;5(8):e003783. doi: 10.1161/JAHA.116.003783 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Bilgrad R National Death Index Plus: Coded Causes of Death: Supplement to the National Death Index User’s Manual. National Center for Health Statistics; 1999. [Google Scholar]
- 33.Xavier AC, Ge Y, Taub JW. Down syndrome and malignancies: a unique clinical relationship. J Mol Diagn. 2009;11(5):371–380. doi: 10.2353/jmoldx.2009.080132 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.McCracken C, Spector LG, Menk JS, et al. Mortality following pediatric congenital heart surgery: an analysis of the causes of death derived from the National Death Index. J Am Heart Assoc. 2018;7(22):e010624.doi: 10.1161/JAHA.118.010624 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin. 2023;73(1):17–48. doi: 10.3322/caac.21763 [DOI] [PubMed] [Google Scholar]
- 36.Cardous-Ubbink MC, Heinen RC, Bakker PJM, et al. Risk of second malignancies in long-term survivors of childhood cancer. Eur J Cancer. 2007;43(2):351–362. doi: 10.1016/j.ejca.2006.10.004 [DOI] [PubMed] [Google Scholar]
- 37.Henderson TO, Whitton J, Stovall M, et al. Secondary sarcomas in childhood cancer survivors: a report from the Childhood Cancer Survivor Study. J Natl Cancer Inst. 2007;99(4):300–308. doi: 10.1093/jnci/djk052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Turcotte LM, Neglia JP, Reulen RC, et al. Risk, risk factors, and surveillance of subsequent malignant neoplasms in survivors of childhood cancer: a review. J Clin Oncol. 2018;36(21):2145–2152. doi: 10.1200/JCO.2017.76.7764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Friedman DL, Whitton J, Leisenring W, et al. Subsequent neoplasms in 5-year survivors of childhood cancer: the Childhood Cancer Survivor Study. J Natl Cancer Inst. 2010;102(14):1083–1095. doi: 10.1093/jnci/djq238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Neglia JP, Friedman DL, Yasui Y, et al. Second malignant neoplasms in five-year survivors of childhood cancer: childhood cancer survivor study. J Natl Cancer Inst. 2001;93(8):618–629. doi: 10.1093/jnci/93.8.618 [DOI] [PubMed] [Google Scholar]
- 41.Im C, Sharafeldin N, Yuan Y, et al. Polygenic risk and chemotherapy-related subsequent malignancies in childhood cancer survivors: a Childhood Cancer Survivor Study and St Jude Lifetime Cohort Study report. J Clin Oncol. 2023;41(27):4381–4393. doi: 10.1200/JCO.23.00428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Asrani SK, Warnes CA, Kamath PS. Hepatocellular carcinoma after the Fontan procedure. N Engl J Med. 2013;368(18):1756–1757. doi: 10.1056/NEJMc1214222 [DOI] [PubMed] [Google Scholar]
- 43.Kuwabara M, Niwa K, Toyoda T, et al. Liver cirrhosis and/or hepatocellular carcinoma occurring late after the fontan procedure—a nationwide survey in Japan. Circ J. 2018;82(4):1155–1160. doi: 10.1253/circj.CJ-17-1053 [DOI] [PubMed] [Google Scholar]
- 44.Kogiso T, Sagawa T, Taniai M, et al. Risk factors for Fontan-associated hepatocellular carcinoma. PLoS One. 2022;17(6):e0270230. doi: 10.1371/journal.pone.0270230 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Lupo PJ, Schraw JM, Desrosiers TA, et al. Association between birth defects and cancer risk among children and adolescents in a population-based assessment of 10 million live births. JAMA Oncol. 2019;5(8):1150–1158. doi: 10.1001/jamaoncol.2019.1215 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Fröhling S, Döhner H. Chromosomal abnormalities in cancer. N Engl J Med. 2008;359(7):722–734. doi: 10.1056/NEJMra0803109 [DOI] [PubMed] [Google Scholar]
- 47.O’Leary L, Hughes-McCormack L, Dunn K, Cooper SA. Early death and causes of death of people with Down syndrome: a systematic review. J Appl Res Intellect Disabil. 2018;31(5):687–708. doi: 10.1111/jar.12446 [DOI] [PubMed] [Google Scholar]
- 48.Irving CA, Chaudhari MP. Cardiovascular abnormalities in Down’s syndrome: spectrum, management and survival over 22 years. Arch Dis Child. 2012;97(4):326–330. doi: 10.1136/adc.2010.210534 [DOI] [PubMed] [Google Scholar]
- 49.Mateos MK, Barbaric D, Byatt SA, Sutton R, Marshall GM. Down syndrome and leukemia: insights into leukemogenesis and translational targets. Transl Pediatr. 2015;4(2):76–92. doi: 10.3978/j.issn.2224-4336.2015.03.03 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Cohen S, Gurvitz MZ, Beauséjour-Ladouceur V, Lawler PR, Therrien J, Marelli AJ. Cancer risk in congenital heart disease—what is the evidence? Can J Cardiol. 2019;35(12):1750–1761. doi: 10.1016/j.cjca.2019.09.023 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.