Infections Among Long-term Survivors of Childhood and Adolescent Cancer: A Report From the Childhood Cancer Survivor Study

Abstract

Purpose

Little is known about infections among adult survivors of childhood cancer. We report the occurrence of and risk factors for infections in a large cohort of survivors of childhood cancer.

Methods

Using the Childhood Cancer Survivor Study (CCSS) cohort, incidence rates of infections among 12,360 five-year survivors of childhood cancer, were compared to those of 4,023 siblings. Infection-related mortality of survivors was compared to the U.S. population. Demographic and treatment variables were analyzed using Poisson regression to determine the rate ratio (RR) and corresponding 95% confidence intervals (CI) for associations with infectious complications.

Results

Compared with the U.S. population, survivors were at an increased risk of death from infectious causes (Standardized Mortality Ratio (SMR)= 4.2; 95% CI, 3.2-5.4), with females (SMR= 3.2; 95% CI, 1.5-6.9) and those exposed to total body irradiation (SMR= 7.8; 95% CI, 1.8-33.0) having the greatest risk. Survivors also reported higher rates than siblings of overall infectious complications (RR=1.3; 95% CI, 1.2-1.4), and higher rates of all categories of infection.

Conclusion

Survivors of childhood cancer remain at elevated risk for developing infectious-related complications, with a higher risk of infection-related mortality years following therapy. Further investigation is needed to provide insight into the mechanisms for the observed excess risks.

Background

Children with cancer have immune dysfunction, as result of their underlying disease, as with lymphoid malignancies, or exposure to chemotherapy.^1-4 Decreased numbers of T-lymphocytes is an important factor in immunodeficiency during maintenance chemotherapy^5,6 However, which other components of the immune system are affected, and to what degree and duration, are unclear.

Infectious complications remain the most important cause of late morbidity and mortality in survivors following hematopoietic cell transplantation (HCT).⁷ It is widely accepted that re-immunization is necessary, and many guidelines have been published.⁸ However, for those children who receive treatment that does not include HCT, risk of late infections and potential need for re-immunization is poorly understood.

Previous studies have shown that by six months after therapy, most patients have recovered immune function, although some patients remain abnormal years later.^9-12 T-lymphocytes may be persistently low, with associated impaired infection protection.^13-14 Survivors have been reported to have lower numbers of lymphocyte subsets than their siblings. The most intensively treated patients may require testing and immunizations beyond six months off therapy.^14-18

Data from the CCSS have demonstrated that long-term survivors experience excess mortality from infectious diseases.¹⁹ The CCSS reported that survivors were 1.6 to 2.7-times more likely to be hospitalized for infection compared to age- and sex-matched individuals in the general population.²⁰ The Canadian Childhood, Adolescent, Young Adult Cancer Survivors Study (CAYACS) similarly reported infections as one of the most common causes of hospitalization and late morbidity.²¹

The present study was designed to investigate the incidence of and risk factors for infection-related disease and mortality in survivors of childhood cancer. This represents the largest most comprehensive investigation of this question to date.

Methods

Inclusion Criteria

The detailed methods of the CCSS cohort have been published previously.^22-24 In brief, the CCSS is a collaborative, multi-institutional project, funded by the National Cancer Institute (Grant No. CA 55727), including individuals who survived five or more years after childhood cancer diagnosis (http://ccss.stjude.org). CCSS is a retrospectively ascertained cohort of 20,346 childhood cancer survivors, and a control group of approximately 4,000 siblings of survivors. CCSS includes 26 participating clinical research centers in the U.S. and Canada, who identified eligible patients with: (1) Diagnosis of leukemia, CNS malignancy (excluding meningioma and craniopharyngioma), Hodgkin or non-Hodgkin lymphoma, neuroblastoma, soft tissue or bone sarcoma, or kidney cancer; (2) diagnosis date between January 1, 1970 and December 31, 1986; (3) age less than 21 years at diagnosis; and (4) alive 5 years from date of diagnosis. The control group includes nearest aged siblings of a random sample of survivors.

Data Collection

The CCSS protocol was reviewed and approved by the Human Subjects Committee at each institution. Data were collected using questionnaires administered in 1994, 2000, 2003 and 2007. A surrogate, typically parent or spouse, was contacted for those eligible subjects who died after achieving 5-year survivorship. The questionnaires addressed social and demographic information, medical conditions, health behaviors, cancer recurrence, development of subsequent neoplasms, and family history. Data collection surveys are available for review at: http://ccss.stjude.org/documents/questionnaires

For all CCSS participants who signed medical release, a detailed summary of exposure to chemotherapy, radiation therapy (RT), and surgery, was abstracted from medical records by trained abstracters. [Information was abstracted regarding exposure to 49 specific chemotherapy agents (cumulative dose for 26). RT data were coordinated through M.D. Anderson Cancer Center's Department of Radiation Physics].²²

Of the 14,358 eligible 5-year survivors who completed the baseline questionnaire, 12,360 survivors were included in the current study, after exclusion of the survivors who developed a subsequent malignant neoplasm (SMN) or had any cancer-related treatment within 12 months before entering the study cohort. The sibling cohort, which consisted of 4,023 siblings who completed the baseline questionnaire, was used for comparison.

Infectious complications were defined by self-report of sinopulmonary infection (bronchitis, sinusitis, tonsillitis, pleurisy, chronic cough, pneumonia), gastrointestinal (GI) infection (hepatitis, and colitis), genitourinary (GU) infection (kidney or bladder infection), and chronic gingivitis. The date of infectious complication was defined as the date of first occurrence. The infectious complication had to have occurred at least 5 years after original cancer diagnosis for incidence analysis, or any time after original cancer diagnosis for cumulative incidence analysis. All patients with infectious complications reported in or before the 2007 follow up questionnaire were included. Infection-related mortality was determined from reviewing ICD-9 and ICD-10 codes associated with death certificates.

Statistical Analysis

Demographic and clinical characteristics [gender, race, age (survivors and siblings), and cancer diagnosis, age at diagnosis and treatment factors (survivors)] were tabulated separately (Table 1). In analyzing occurrence of infection-related complications, death, late relapse (5 years past diagnosis) and SMN were treated as competing risk events, since these would change the rates of infection-related events, and complete information on all relapses and SMNs is unavailable.

Table 1. Characteristics of Study Population.

Characteristic	Survivors(N=12,360)		Siblings(N=4,023)
	N	%	N	%
Gender
Male	6564	53.1	1937	48.1
Female	5796	46.9	2086	51.9
Race
White	10634	86	3509	87.2
Black	637	5.2	112	2.8
Hispanic	650	5.3	148	3.7
Other/missing	439	3.6	254	6.3
Age at latest questionnaire
<20 years	1058	8.6	267	6.6
20-29 years	3932	31.8	1134	28.2
30-39 years	4632	37.5	1392	34.6
40+ years	2738	22.2	1230	30.6
Cancer diagnosis
Acute Lymphoblastic Leukemia	3279	26.5
Hodgkin Lymphoma	1733	14
Kidney Tumors	1200	9.7
Soft Tissue Sarcoma	1135	9.2
Astrocytomas	1087	8.8
Non-Hodgkin Lymphoma	1001	8.1
Neuroblastoma	889	7.2
Osteosarcoma	647	5.2
Medulloblastoma, PNET	338	2.7
Ewing Sarcoma	324	2.6
Acute Myelogenous Leukemia	290	2.3
Other CNS Tumors	283	2.3
Other Leukemia	105	0.8
Other Bone Tumors	49	0.4
Age at diagnosis
<1 years	932	7.5
1-3 years	3059	24.7
4-7 years	2713	21.9
8-10 years	1331	10.8
11-14 years	2131	17.2
15-20 years	2194	17.8
Splenectomy
Yes	1091	8.8
No	9482	76.7
Not Available+	1787	14.5
Abdominal RT
Yes	2685	21.7
No	7999	64.7
Not Available+	1676	13.6
Chest RT
Yes	2781	22.5
No	7903	63.9
Not Available+	1676	13.6
Total Body Irradiation
Yes	137	1.1
No	10547	85.3
Not Available+	1676	13.6
Chemotherapy
Yes	8180	66.2
No	2405	19.5
Not Available+	1775	14.4
Steroids
Yes	4726	38.2
No	5859	47.4
Not Available+	1775	14.4

⁺Participants did not provide permission for release of treatment information, or the information is missing.

To describe infectious complications, cumulative incidences individually and combined were calculated for: sinopulmonary, GI and GU infection, and gingivitis. Infectious complications that occurred within 5 years from cancer diagnosis were included as prevalent cases at the study-cohort entry.

Our rate analysis included: (1) analysis at the study-cohort entry of having developed an infection-related complication within 5 years from the cancer diagnosis (prevalence analysis at the study-cohort entry); and (2) analysis of developing an infection-related complication from the study-cohort entry (incidence analysis after the study-cohort entry). The incidence analysis is not tenable for the period from cancer diagnosis to study-cohort entry because the survivors had to survive 5 years to enter the cohort. For the prevalence analysis, the prevalence ratios comparing survivors with siblings were individually estimated for all of the specific infectious categories. For the incidence analysis, incidence rates were estimated by dividing the observed counts of incident cases by the person-years at risk, starting at 5 years post- cancer diagnosis, to the earliest of the specific infection-related complication of interest, the end of follow up, or any of the competing-risk events. The relative rates comparing survivors with siblings were calculated using Poisson regression, adjusting for age during follow-up (time dependent), sex, and race. Potential within-family correlation was accounted for by using the generalized estimating equation. ²⁵

Among survivors, multivariable regression analyses were performed to evaluate effects of demographic and treatment variables on sinopulmonary, GI and GU infection, and gingivitis, individually and combined, with covariates (sex, cancer diagnosis, age at diagnosis, splenectomy, abdominal RT, chest RT, TBI, chemotherapy and steroids) included by backward selection. Analysis of prevalence was performed using log-binomial regression, adjusting for age at the time of study-cohort entry, and for the incidence analysis using Poisson regression, adjusting for age during follow-up (time dependent).

Multivariable Poisson regression was also performed to evaluate effects of demographic and treatment variables on infection-related standardized mortality ratios (SMR) using the age-sex-, and calendar year-specific U.S. population rates (http://www.cdc.gov/) using the same method of covariate selection as described above.

For missing dates of self-reported infections, we used the multiple imputation method, proposed by Taylor et al.²⁶, to impute the missing times of disease onset. For each of the 10 types of infections, we used all potential covariates in the regression analyses to impute the missing time of the onset of the reported infectious event. Ten multiply imputed datasets were analyzed and summarized by the standard method for combining multiple-imputation ²⁷.

Results

The characteristics of the study population are listed in Table 1. Approximately 53% of survivors were male (48% of siblings), 86% were Caucasian (87% of siblings), and 22% were over age 40 years at time of last follow-up (31% of siblings). Of the survivors, 27% have a history of acute lymphoblastic leukemia (ALL), and 65% were 10 years of age or less at diagnosis. With regard to treatment history, 66% of survivors have a history of exposure to chemotherapy, 38% received steroid therapy, 23% underwent chest radiation, 22% abdominal radiation, 9% splenectomy, and 1% total body irradiation (TBI).

Incidence rates of infections are provided in Table 2. When compared to siblings and after adjusting for sex, age and race, there was a statistically significantly higher rate of infections among survivors (RR= 1.3; 95% CI 1.2-1.4). Survivors reported higher rates of all categories of infection, most notably pneumonia (RR= 3.7; 95% CI, 2.6-5.3), hepatitis (RR= 2.5; 95% CI, 2.0-3.3), and sinusitis (RR=1.8;95% CI, 1.6-1.9).

Table 2. Incidence Rates of Infectious Complications.

Infectious complications		Reported Outcome		From cancer diagnosis to 5 Years			5+ years after cancer diagnosis
		Survivors N=12,360	Siblings N=4,023	N	Prevalence Ratio	95% CI†	N	Rate*	95%CI†	Rate Ratio **	95%CI†
Sinopulmonary Infection	Bronchitis	2953	1201	1037	1.4	1.3 - 1.5	1916	14.3	13.7 - 15.0	1.4	1.3 - 1.5
	Sinusitis	2509	760	836	2.0	1.8 - 2.2	1673	11.8	11.2 - 12.3	1.8	1.6 - 1.9
	Tonsillitis	1789	1151	807	0.8	0.8 - 0.9	982	7.3	6.9 - 7.8	1.1	1.0 - 1.2
	Pleurisy	414	119	104	2.8	2.0 - 3.9	310	1.8	1.6 - 2.1	1.4	1.1 - 1.7
	Chronic Cough	1327	273	436	4.0	3.3 - 4.9	891	4.3	4.0 - 4.6	2.0	1.8 - 2.3
	Pneumonia	369	43	144	9.5	5.8 - 15.5	225	1.0	0.9 - 1.2	3.7	2.6 - 5.3
	Any event	4804	2179	2118	1.3	1.2 - 1.4	2686	22.7	21.8 - 23.6	1.3	1.2 - 1.4
	More than one event	2860	941	888	1.7	1.5 - 1.9	1972	11.1	10.6 - 11.6	1.5	1.4 - 1.7
Gastrointestinal Infection	Hepatitis	576	78	252	6.5	4.8 - 9.0	324	1.5	1.3 - 1.7	2.5	2.0 - 3.3
	Colitis	130	53	29	1.1	0.7 - 1.9	101	0.8	0.7 - 1.0	1.2	0.9 - 1.7
	Any event	696	129	277	4.6	3.5 - 5.9	419	2.0	1.8 - 2.2	1.9	1.5 - 2.3
	More than one event	11	2	4	3.3	0.6 - 17.6	7	0.0	0.0 - 0.1	2.0	0.4 - 10.6
Genitourinary Infection	Kidney or Bladder infection	1278	502	398	1.4	1.2 - 1.6	880	4.6	4.3 - 4.9	1.2	1.1 - 1.4
Chronic Gingivitis		664	196	72	2.0	1.3 - 3.0	592	4.1	3.8 - 4.5	1.6	1.4 - 1.9
Any of the above infections		5578	2436	2470	1.4	1.3 - 1.4	3108	28.8	27.7 - 29.8	1.3	1.2 - 1.4

^*rate: per 1000 person year

^**Relative to sibling cohort adjusted for sex, age, and race

^†CI: confidence interval

Results of multivariable analyses of demographic and cancer treatment factors associated with infectious complications are shown in Table 3. Factors associated with higher rates of overall late infectious complications included: female sex (RR= 1.7; 95% CI, 1.6-1.8), diagnosis of Hodgkin lymphoma (RR= 1.3; 95% CI, 1.2-1.6), and older age at cancer diagnosis [using 15-20 years of age as the referent group: <1 year of age (RR= 0.6; 95% CI, 0.5-0.8); 1-3 years of age (RR= 0.6; 95% CI, 0.5-0.8); 4-7 years of age (RR= 0.8; 95% CI, 0.6-0.9)]. Patients who reported being a current (RR=1.4; 95% CI, 1.2-1.5) or former (RR=1.3; 95% CI, 1.1-1.5) tobacco smoker had elevated rates of sinopulmonary infection compared with those who reported having never smoked.

Table 3. Multivariable Analysis of Infection-Related Complications by Demographic and Treatment Factors.

	Sino-pulmonary Infection		Gastrointestinal Infection		Genitourinary Infection		Gingivitis		Any infection
	Prevalence Ratio+ (95%CI†)	Rate Ratio++ (95%CI)	Prevalence Ratio (95%CI)	Rate Ratio (95%CI)	Prevalence Ratio (95%CI)	Rate Ratio (95%CI)	Prevalence Ratio (95%CI)	Rate Ratio (95%CI)	Prevalence Ratio (95%CI)	Rate Ratio (95%CI)
Gender
Male (ref)
Female	1.2(1.1 - 1.4)*	1.8(1.6 - 1.9)*	1.4(1.0 - 1.8)*		4.3(3.3 - 5.8)*	6.7(5.5 - 8.1)*		1.2(1.0 - 1.4)	1.3(1.2 - 1.4)*	1.7(1.6 - 1.8)*
Smoke
Current	1.1(0.9 - 1.2)	1.4(1.2 - 1.5)*
Former	0.9(0.8 - 1.1)	1.3(1.1 - 1.5)*
Never
Cancer diagnosis
Acute lymphoblastic leukemia(ref)
Acute myeloid leukemia	1.1(0.9 - 1.5)	1.2(0.9 - 1.7)	2.4(1.6 - 3.5)*	1.6(0.9 - 2.6)	0.8(0.2 - 2.4)	1.0(0.6 - 1.7)	1.1(0.2 - 8.2)		1.3(1.0 - 1.6)*	1.1(0.9 - 1.5)
Other leukemia	0.9(0.5 - 1.4)	1.0(0.6 - 1.8)	0.2(0.0 - 1.7)	1.3(0.5 - 3.1)	1.1(0.3 - 4.1)	2.1(1.1 - 4.2)*	5.6(1.3 - 24.0)*		0.8(0.5 - 1.3)	1.4(1.0 - 2.0)
Astrocytomas	0.8(0.7 - 1.1)	0.9(0.7 - 1.1)	0.2(0.1 - 0.6)*	0.4(0.2 - 0.9)*	3.3(1.9 - 5.5)*	1.1(0.8 - 1.4)	4.7(1.6 - 14.0)*		0.9(0.7 - 1.1)	0.8(0.7 - 1.0)*
Medulloblastoma, PNET	0.8(0.5 - 1.1)	0.9(0.6 - 1.2)	0.1(0.0 - 0.6)*	0.5(0.2 - 1.2)	0.9(0.3 - 3.0)	0.6(0.3 - 1.2)	3.1(0.6 - 15.6)		0.8(0.6 - 1.0)	0.9(0.7 - 1.2)
Other CNS tumors	0.9(0.6 - 1.3)	0.7(0.5 - 1.1)	0.0(0.0 - 0.2)*	0.9(0.4 - 2.0)	2.4(1.0 - 5.4)*	1.0(0.6 - 1.8)	7.4(1.9 - 28.6)*		0.8(0.6 - 1.2)	0.8(0.6 - 1.0)
Hodgkin Lymphoma	1.2(0.9 - 1.5)	1.6(1.3 - 1.9)*	0.2(0.1 - 0.4)*	0.5(0.3 - 0.7)*	1.0(0.6 - 1.8)	1.2(0.9 - 1.5)	0.9(0.3 - 2.5)		1.0(0.8 - 1.2)	1.3(1.2 - 1.6)*
Non-Hodgkins lymphoma	1.2(1.0 - 1.4)	1.3(1.1 - 1.6)*	0.5(0.3 - 0.8)*	0.5(0.3 - 0.8)*	2.3(1.4 - 4.0)*	1.4(1.0 - 1.9)*	0.8(0.2 - 3.8)		1.1(0.9 - 1.3)	1.2(1.0 - 1.4)*
Kidney tumors	0.9(0.7 - 1.1)	1.2(1.0 - 1.4)*	0.4(0.2 - 0.7)*	0.6(0.4 - 1.0)*	2.2(1.4 - 3.4)*	1.9(1.5 - 2.4)*	0.4(0.1 - 3.1)		0.9(0.8 - 1.1)	1.1(1.0 - 1.3)*
Neuroblastoma	1.0(0.8 - 1.2)	1.2(1.0 - 1.5)	0.2(0.1 - 0.6)*	0.6(0.3 - 1.1)	3.1(1.9 - 5.2)*	1.4(1.0 - 1.9)	0.0(0.0 - 2.5)		1.0(0.8 - 1.2)	1.1(0.9 - 1.3)
Soft tissue sarcoma	1.0(0.8 - 1.1)	1.0(0.8 - 1.2)	0.4(0.2 - 0.6)*	0.7(0.5 - 1.1)	4.1(2.7 - 6.3)*	1.8(1.4 - 2.2)*	3.1(1.4 - 7.2)*		1.1(0.9 - 1.2)	1.1(0.9 - 1.3)
Ewings sarcoma	0.9(0.6 - 1.2)	1.5(1.1 - 2.0)*	0.3(0.1 - 0.7)*	0.6(0.3 - 1.2)	2.6(1.3 - 5.0)*	1.1(0.6 - 1.8)	1.3(0.3 - 6.5)		0.9(0.7 - 1.2)	1.3(1.0 - 1.6)
Osteosarcoma	0.8(0.6 - 1.0)	1.2(0.9 - 1.6)	0.4(0.2 - 0.8)*	0.9(0.6 - 1.5)	1.6(0.7 - 3.4)	1.1(0.7 - 1.6)	1.0(0.2 - 4.6)		0.7(0.6 - 1.0)*	1.1(0.9 - 1.4)
Other sarcoma	1.0(0.4 - 2.6)	0.6(0.2 - 1.9)	0.0(0.0 - 1.1)	1.2(0.3 - 5.3)	0.0(0.0 - 3.0)	1.5(0.5 - 4.2)	0.0(0.0 - 12.3)		0.9(0.4 - 2.1)	0.7(0.3 - 1.6)
Age at diagnosis
<1 year		0.5(0.4 - 0.6)*		0.5(0.2 - 0.9)*		0.7(0.5 - 1.0)		1.7(1.0 - 2.8)*		0.6(0.5 - 0.8)*
1-3 years		0.5(0.4 - 0.6)*		0.5(0.3 - 0.8)*		0.6(0.4 - 0.8)*		1.8(1.3 - 2.6)*		0.6(0.5 - 0.8)*
4-7 years		0.6(0.5 - 0.7)*		0.5(0.3 - 0.7)*		0.7(0.5 - 0.9)*		1.6(1.2 - 2.2)*		0.8(0.6 - 0.9)*
8-10 years		0.8(0.6 - 1.0)*		0.7(0.5 - 1.1)		0.7(0.5 - 1.0)*		1.4(1.0 - 2.0)*		0.8(0.7 - 1.0)
11-14 years		0.8(0.7 - 1.0)		0.6(0.4 - 0.9)*		0.9(0.7 - 1.1)		1.4(1.1 - 1.9)*		1.0(0.8 - 1.1)
15-20
Splenectomy
Yes
No (ref)
Abdominal RT
Yes	0.8(0.7 - 1.0)*		1.6(1.2 - 2.1)*		1.5(1.1 - 2.0)*
No (ref)
Chest RT
Yes	1.1(1.0 - 1.3)
No(ref)
Total Body Irradiation
Yes	1.5(1.1 - 2.2)*			2.1(1.1 - 4.0)*					1.4(1.1 - 1.8)*
No (ref)
Chemotherapy
Yes	1.2(1.0 - 1.4)		1.9(0.9 - 3.8)	1.7(1.1 - 2.5)*	1.5(1.1 - 2.1)*		2.7(1.2 - 6.5)*		1.2(1.1 - 1.4)*
No (ref)
Steroids
Yes								1.3(1.1 - 1.6)*
No(ref)

^*P value<0.05

^†CI: confidence interval

⁺Prevalence for events present prior to cohort entry

⁺⁺Incidence for events occurring after entry into cohort

History of TBI or chemotherapy was significantly associated with increased rates of late gastrointestinal infections. Survivors who received steroids had an increased rate of chronic gingivitis (RR= 1.3; 95% CI, 1.1-1.6). No other treatment variables, including splenectomy, abdominal radiation, or chest radiation were found to be associated with a statistically significant increase in late infections. There was no significant difference in patterns of late infections and their associations with various risk factors when we removed survivors who received BMT.

Cancer diagnosis and demographic factors were independently associated with an increased incidence for many of the specific categories of late infections investigated (Table 3). Using childhood ALL as the referent population, increased risks for sinopulmonary infection were present in survivors of Hodgkin lymphoma, non-Hodgkin lymphoma, kidney tumors, and Ewing sarcoma. Increased risks of genitourinary infections were present for survivors of non ALL/AML leukemia, non-Hodgkin lymphoma, kidney tumors, and soft-tissue sarcomas. Conversely, survivors of childhood ALL had significantly higher infection rates than survivors of astrocytoma (sinopulmonary, GI), Hodgkin lymphoma (GI), non-Hodgkin lymphoma (GI), and kidney tumors (GI). Increased risks were found for female sex (sinopulmonary, GU, gingivitis), and age 15-20 years at cancer diagnosis (sinopulmonary, GI, and GU infections).

Cumulative incidence of infection-related complications by category of infection, and of infectious-related mortality are shown in Figures 1 and 2, respectively. Compared with the U.S. population, survivors had an overall increased rate of death from infectious causes (SMR=4.2; 95% CI,3.2-5.4), with females (SMR=3.2; 95% CI, 1.5-6.9) and those exposed to TBI (SMR=7.8; 95% CI, 1.8-33.0) having the highest mortality. Among the 65 deaths attributed to an infectious cause, 25% were due to pneumonia, 17% to septicemia, 12% to HIV-associated infections, 9% to bacterial endocarditis, and 37% to miscellaneous other causes (e.g., gastrointestinal infections, encephalitis, etc.).

Figure 1.

Cumulative incidence of infection-related complications by category of infection, and of infectious-related mortality.

Figure 2.

Compared to siblings, survivors of childhood cancer have higher rates of infectious complications. In comparison to the U.S. population, survivors are at an increased risk of death from infections.

Discussion

To our knowledge, the present study is the largest to date to explore late infectious complications in childhood cancer survivors. In our cohort, nearly one-third of survivors reported infectious complications at greater than five years from diagnosis. Twenty-two percent of survivors reported late sinopulmonary infections, including 16% with multiple recurrences. Other types of infection were less common, although not inconsequential (genitourinary 7%, gastrointestinal 3%, chronic gingivitis 5%). For all categories of late infection, survivors had a significantly higher rate compared to siblings.

In the multivariable analysis, Hodgkin and non-Hodgkin lymphoma survivors had the highest significant risk of late overall infections. Survivors of all other disease categories, expect for survivors of CNS cancer, were at a higher risk for at least one type of infection compared to survivors of ALL. Many CNS tumor patients received surgery alone, which may in part explain this group's lower rates of infections.

Survivors of childhood AML are a group purported to be at greater risk for late infections due to potential long-term immune deficiencies.²⁸ In the present investigation, while there was suggestion of a modestly higher risk of infections in survivors of AML, these differences were not statistically significant.

With regard to risk of infectious complications by treatment, functional (i.e. radiation-induced) or anatomic (i.e. surgical splenectomy) asplenia predisposes childhood cancer survivors to a lifelong risk of sepsis.²⁹ In the present study with multivariable analysis, asplenia was not associated with an increased risk of late infectious complications. This may be due to the institution of preventative measures in asplenic survivors, such as immunizations and prophylactic antibiotics. However, abdominal radiation was associated with higher risk of infection. The degree of functional asplenia these survivors sustain is less defined than with splenectomy, so they may not be receiving the same level of preventative care.

Infectious complications remain the leading cause of late morbidity and mortality in survivors of HCT^7.8. In the present evaluation, HCT survivors treated with TBI were found to be at higher risk for late GI infections and infectious-related mortality than those without history of TBI. The overall numbers of TBI-exposed survivors in this cohort are relatively small, and information on graft-versus-host-disease is unknown, so it is possible that other potential differences were not evident. Other than an elevated incidence of chronic gingivitis in survivors who received steroid therapy, no other individual treatment variable was found to be associated with an increased risk for late infectious complications.

In this investigation, female survivors had a significantly higher rate of late infections, particularly GU, with nearly seven times the relative rate of that observed in males. Females also had a greater than three-fold higher rate of death from infections. The reasons for these gender differences are unclear, and require further study.

Older age at cancer diagnosis was associated with a higher incidence of late infections, except for gingivitis. It is possible that immune reconstitution following chemotherapy is more robust in younger children whose immune repertoire is still developing.

From a prevention standpoint, other than for survivors with history of HCT, there are no standard guidelines for re-immunizing or testing titers. There are also no widely accepted guidelines for prophylactic immunizations, other than encapsulated organism vaccines for asplenic patients. There is also ongoing debate regarding the use of prophylactic antibiotics for asplenic patients. ^29,30 The Working Party of the British Committee for Standards in Clinic Hematology Task Force recommends lifelong antibiotics,³¹ the American Academy of Pediatrics recommends antibiotics until age 5,³² and Children's Oncology Group's Survivorship Guidelines recommend consideration of antibiotics.³³ Better defined guidelines for re-immunization, prophylactic immunization and antibiotic use are needed.

The strengths of this study include the large number of survivors followed into adulthood, the availability of detailed treatment information, and our use of a sibling comparison group. However, several limitations need to be considered when interpreting the results. First, relying on self-reported data could result in both over and underreporting of certain outcomes, thus either differential or non-differential misclassification. Cancer survivors may have a heightened level of awareness of symptoms of illness as a result of their cancer experience, including higher rates of anxiety, depression or somatization, which could result in reporting bias when compared to siblings. However, the significantly increased risks of infections observed in the current study, remained significant when we restricted analyses to survivors without significant emotional distress (as assessed using the Behavioral Symptoms Inventory-18). Furthermore, the current study relied upon self-report data, and thus we were not able to accurately ascertain severity of infection, e.g. whether a mild illness involving sinus congestion vs. a severe sinusitis requiring parenteral antibiotics. However, we are unaware of data suggesting variability in symptom-definition and/or reporting according to symptom severity would differ between patients and siblings, or across subgroups of survivors.

Infection-related mortality utilized death certificates, which may be associated with some degree of non-differential misclassification. While current therapy for the majority of pediatric cancer patients includes the same agents used for the CCSS cohort,^34,35 treatment protocols change over time, employing less or more intensive regimens. Thus, results from the current study may not be applicable to individuals treated on more contemporary protocols. The CCSS is currently undergoing expansion to include those patients diagnosed between 1987 and 1999. Future analyses will be necessary to compare the present study's results to more modern treatment regimens to determine if, for example, with elimination of routine splenectomy and decreased radiation doses, a lower incidence of infectious complications is observed.

In conclusion, long-term survivors of childhood cancer are at an increased risk of infections and infectious-related mortality, and this risk continues to increase with time from diagnosis. Females, those treated for Hodgkin and non-Hodgkin lymphoma, and those over 15 years of age at diagnosis appear to be at highest risk. Females and those who received TBI are at the greatest risk of infectious-related mortality. Longitudinal assessment of immunologic function of survivors of childhood cancer may provide further insight into the prevalence and pathogenesis of abnormal immunologic parameters that may place survivors at increased risk for infection. Given the observed excess risk of infection-related mortality, survivors and their clinical care providers should be aware of the need for appropriate immunizations and prompt care for suspected infection.