Childhood Cancer Casts Long Shadow Over Adult Health

NEW YORK, Oct. 12 -- Survivors of childhood cancer may face futures fraught with vastly greater chronic and life-threatening health risks than those of their siblings, according to a large retrospective study.

Childhood cancer survivors of the 1970s and 1980s were more than three times as likely as their siblings to have a chronic health condition and more than eight times as likely to have a severe or life-threatening condition in the next five to 30 years, investigators reported in the Oct. 12 issue of the New England Journal of Medicine.

Among 10,397 survivors in the study, 62.3% had at least one chronic condition and 27.5% had a severe or life-threatening condition, the researchers found.

Furthermore, 37.6% of the survivors reported having at least two conditions, and 23.8% reported three or more conditions. In contrast, 36.8% of 3,034 siblings in the study reported having a single chronic health condition, of which the condition was serious for only 5.2%.

Major secondary risks included second cancers, cardiovascular disease, renal dysfunction, severe musculoskeletal problems, and endocrinopathies, said Kevin Oeffinger, M.D., of Memorial Sloan-Kettering Cancer Center here, and Anna Meadows, M.D., of Children's Hospital of Philadelphia, a pioneer in the field, and other members of the Childhood Cancer Survivor Study.

On the whole, long-term survivors were particularly vulnerable to a lengthy list of severe or life-threatening or disabling conditions.

These included major joint replacement not part of original cancer therapy, congestive heart failure, second malignant neoplasm (such as breast and colorectal cancer), melanoma, severe cognitive dysfunction, coronary artery disease, cerebrovascular accident, renal failure or dialysis, musculoskeletal problems, endocrinopathies (premature gonadal failure, thyroid disease, osteoporosis, and hypothalamic and pituitary dysfunction), hearing loss, legal blindness or loss of an eye, and ovarian failure in women.

The relative risk and 95% confidence intervals for severe or life-threatening or disabling health conditions were:

• Major joint replacement, 54.0 (7.6-386.3)
• Congestive heart failure, 15.1 (4.8-47.9)
• Second malignant neoplasm 14.8 (7.2-30.4)
• Cognitive dysfunction, severe 10.5 (2.6-43.0)
• Cerebrovascular accident 9.3 (4.1-25.9)
• Renal failure or dialysis 8.9 (2.2-36.6)

The results came from 26 geographically diverse collaborating institutions in the Childhood Cancer Survivor Study. The CCSS is a retrospective cohort study that tracked the health of adults who received a diagnosis of childhood cancer between 1970 and 1986 and compared the results with those for siblings.

Those in the study were younger than 21 at diagnosis and survived for at least five years after that. At the time of the study, mean age of the survivors was 26.6 years (range, 18 to 48); mean age of the siblings was 29.2 years (range, 18 to 56). Of the survivors, 46.2% were women.

The study was restricted to participants who had an original diagnosis of leukemia, central nervous system tumor, Hodgkin's disease, non-Hodgkin's lymphoma, Wilms' (kidney) tumor, neuroblastoma, soft-tissue sarcoma, or bone tumor.

Severity scores for each new condition ranged from mild (grade one), moderate (grade two), and severe (grade three) to life-threatening or disabling (grade four), and fatal (grade 5).

The adjusted relative risk (RR) of a chronic condition in a survivor, compared with siblings, was 3.3 (95% CI, 3.0 to 3.5) and 8.2 (CI, 6.9 to 9.7) for a severe or life-threatening condition. Survivors were 4.9 times as likely to have two or more chronic health conditions (CI, 4.4-5.5).

Groups at highest risk for having a severe or disabling late condition were survivors of bone tumors (RR 38.9, CI, 31.2 to 48.5), CNS tumors (RR, 12.6, CI 10.3 to 15.5), and Hodgkin's disease (RR 10.2, CI 8.3 to 12.5).

Hodgkin's disease survivors had the highest risk of second cancers and heart disease, the researchers wrote.

All cancer groups were significantly more likely to have any condition, a severe or life-threatening condition, and two or more conditions, compared with siblings (P<0.001 for all comparisons), the researchers reported.

Serious health risks were greater for women. Compared with male survivors, after adjustment for type of cancer, age at the time of the study, and race or ethnic group, female survivors were 1.5 times (CI, 1.3 to 1.5) as likely to have any serious (grades three or four) condition. They were also more likely to have multiple conditions (RR, 1.5, CI, 1.4 to 1.7).

Overall, women were reported to have a greater risk of diminished health status, second cancers, anthracycline-related cardiomyopathy and congestive heart failure, cognitive dysfunction related to cranial radiotherapy-related, growth hormone deficiency, and obesity.

The age of the survivor at diagnosis also independently modified the risk after adjustment for confounding factors. For each cancer group, survivors who received the diagnosis at an older age were significantly more like to report health consequences. However, for the most part, race or ethnic group did not affect the likelihood of a serious late condition, although black-non-H.

ispanic survivors were less likely than white non-Hispanic survivors to have any condition.

Certain therapy combinations led to a tenfold increase in the relative risk of a serious condition (grades three or four). These were chest radiation plus Blenoxane (bleomycin), chest radiation plus an anthracycline, chest radiation plus abdominal or pelvic irradiation, an anthracycline plus an alkylating agent, and abdominal or pelvic irradiation plus an alkylating agent.

An increase in the cumulative dose of an alkylating agent was associated with an increased risk of any or multiple conditions. In a similar way, an increase in the cumulative dose of an alkylating agent in combination with any type of irradiation was associated with an increased relative risk. In contrast, the researchers said, an increase in the total dose of an anthracycline was not associated with an increased risk of any condition.

There are several considerations to consider when interpreting these findings, the researchers noted. First, the conditions were self-reported without external verification, except for death. Also, several key chronic conditions, such as late-onset cardiomyopathy, associated with previous anthracycline exposure, may remain clinically silent for long periods.

Other conditions that may be under-reported are osteoporosis, hypertension, and insulin resistance. A notable omission in the comprehensive list of chronic conditions in this study is adverse mental health outcomes, an important component of morbidity after childhood cancer.

In summary, the researchers wrote, adult survivors of pediatric cancer who were treated in the 1970s and 1980s are a high-risk population. Thirty years after diagnosis, almost three-fourths of the survivors had a chronic health condition, more than 40% had a serious health problem, and one-third had multiple conditions. The rate of health conditions in this population increased with time and did not appear to plateau, they said.

Yet the results of pediatric oncology trials have led to improvements in treatment over the years, Dr. Meadows said. She predicted that children diagnosed with cancer since 1987 will have fewer chronic problems as survivors than the cohort analyzed in this study.

One implication of this study, Drs. Oeffinger and Meadows wrote, is the need to continue medical surveillance of adult survivors of childhood cancer. There are 27,000 survivors of pediatric cancer in the U.S., yet fewer than 20% of these survivors of childhood cancer are followed at a cancer center or by an oncologist, and the likelihood of follow-up in this setting decreases even more with time.

"For this reason," the authors said, "it is important for general physicians and internists to be aware of the risks facing this population."