Grading of late effects in young adult survivors of childhood cancer followed in an ambulatory adult setting
The objective of the current study was to describe a multidisciplinary transition program for following young adult survivors of childhood cancer in an adult-based ambulatory medical setting and to report the late effects with grades of toxicity diagnosed in all adult survivors followed in the program.
The study population was comprised of all young adult survivors (n = 96) of childhood cancer who were seen in the After the Cancer Experience (ACE) Young Adult Program prior to January 31, 1999. The median age of the survivors was 22.8 years (range, 17–34 years) and the median interval from the time of cancer diagnosis was 15.2 years (range, 6–25 years). Primary cancer groups included: leukemia, 33%; sarcoma, 24%; Hodgkin disease, 15%; non-Hodgkin lymphoma, 12%; Wilms' tumor, 9%; and other, 7%. Late effects were graded using the Common Toxicity Criteria, Version 2 (CTCv2), developed by the National Cancer Institute.
Approximately 69% of the patients (66 of 96) had at least 1 late effect. Thirty-three percent of patients had a single late effect whereas 36% had ≥ 2 late effects. Thirty percent of patients had a CTCv2 Grade 3 or 4 late effect.
The current study represents an example of a successful multidisciplinary transition program in an ambulatory, adult setting for young adult survivors of childhood cancer. Late effects of cancer treatment are common in young adult survivors, with approximately 33% being moderate to severe. Further studies are needed to modify CTCv2 with the aim of developing a reliable and valid tool to assess late effects in long term survivors of childhood cancer. Cancer 2000;88:1687–1695. © 2000 American Cancer Society.
As the number of childhood cancer survivors grows, attention is increasingly being directed to the evaluation and management of late effects or sequelae of treatment.1 Late effects range from relatively benign conditions such as radiation-induced alopecia to potentially life-threatening problems such as radiation-related breast carcinoma or anthracycline-induced cardiomyopathy.2–5 Virtually all organ systems can be affected by either radiation, chemotherapy, or surgery. Because our understanding of the natural history of late effects is evolving slowly, it is important to maintain contact with and periodically study this population of patients. Regularly scheduled surveillance, with early detection and treatment of late effects, combined with education concerning risk modification theoretically should impact the quality of life and long term health of adult survivors and should be investigated further.
Most adult survivors of childhood cancer are not followed on a regular basis. Preliminary data from the National Institutes of Health funded, multicenter Childhood Cancer Survivor Study show that 51% of approximately 14,000 responding long term survivors of childhood cancer had not been seen by a physician during the previous 2 years for evaluation of cancer-related problems.6 A recent survey of the members of the Pediatric Oncology Group and Children's Cancer Group reported that 44% of responding institutions have a mechanism for following adult survivors, but only 15% of the programs have established a formal database for this population.6
Multidisciplinary transition programs, combining the expertise of pediatric oncologists with health care providers experienced in the health needs of adult patients, have been recommended as a possible mechanism for the long term follow-up of childhood cancer survivors.7–11 However, to our knowledge few programs exist that combine these areas of expertise in the care of adult survivors of childhood cancer.6
The purpose of the current study was to: 1) describe a multidisciplinary transition program for following young adult survivors of childhood cancer in an adult-based ambulatory medical setting and 2) to report the late effects with grades of toxicity diagnosed in the first 96 adult survivors who were followed in the program.
MATERIALS AND METHODS
After the Cancer Experience (ACE) Young Adult Program
The After the Cancer Experience (ACE) Young Adult Program was developed in 1995 as a collaborative effort between the Center for Cancer and Blood Disorders at Children's Medical Center of Dallas and the Department of Family Practice and Community Medicine at the University of Texas Southwestern. The program was developed as an extension of the ACE Program, a late effects program that started in 1989 and follows childhood cancer survivors age < 18 years who are at least 24 months posttreatment. The ACE Program collaborates with the Cancer Registry of the Children's Medical Center of Dallas, begun in 1990, in tracking the status of childhood cancer survivors treated at the institution.
Adult survivors ≥ 18 years of age are eligible for regular follow-up in the ACE Young Adult Program. Patients are followed annually and examined and screened for late effects of cancer therapy, second malignant neoplasms, and risk factors that may contribute to the early onset of common adult health problems. Protocols used for routine testing of asymptomatic young adult survivors of childhood cancer seen in the ACE Young Adult Program, adapted from current literature and modified for the patient population, are described in Table 1. Additional laboratory, radiographic, or other testing is performed as indicated, based on the clinical history or findings. Patient education materials, verbal and written, have been collected or developed within the program and are provided for areas of concern and health risk reduction. Late effects diagnosed at each visit are maintained in a formal database, which includes patient demographics and treatment protocols.
Table 1. Treatment Specific Protocol for the Routine Screening of Asymptomatic Survivors Seen in the After the Cancer Experience (ACE) Young Adult Program
|If patient received:|
| Actinomycin or antimetabolite||ALT||Periodically|
| Aminoglycoside, high dose||Audiology||Optional|
| Anthracycline||Echocardiogram||Every 3 years|
| (≥ 300 mg/m2, or anthracycline administered prior to age 1 one year or ≥ 200 mg/m2 with radiation involving the chest)||EKG||Optional|
| BCNU, CCNU, bleomycin||CXR||Baseline|
|Pulmonary function tests||Baseline and as needed|
| Cisplatin||BUN, creatinine, magnesium||Annually|
| Corticosteroids||Bone densitometry||Optional|
| Cyclophosphamide||FSH, LH, estradiol||Optional|
| Cyclosporine||Bone densitometry||Optional|
| Etoposide||CBC with platelets and differential||Annually|
| Nitrogen mustard, procarbazine||CBC with platelets and differential||Annually|
|FSH, LH, estradiol||Optional|
|If patient received:|
| Cranial or craniospinal radiation||Cataract screening||Periodically|
|TSH, Free T4||Annually|
| Mantle radiation||TSH||Annually|
|Mammogram (females)||Start 8 years after radiation, then annually|
|Plain radiographs of irradiated site||Optional|
| Abdominal radiation||Hemoccult screening||Annually|
| Pelvic radiation||FSH, LH||Optional|
| High dose radiation of the trunk or extremities||Plain radiographs of the irradiated sites||Optional|
|If patient received:|
| Nephrectomy||BUN, creatinine, urinalysis||Annually|
| Splenectomy||Verify immunizations||Annually|
|If patient received blood products before July 1992||HIV||Baseline|
The ACE Young Adult Program, centered in an adult-based ambulatory care center, is directed by a family practice physician (K.O.) and coordinated by a pediatric oncology nurse practitioner (D.E.). Also included on the ACE Young Adult team are two pediatric oncologists and two other family physicians. A consulting network of adult and pediatric subspecialists has been developed by the team and includes cardiologists, endocrinologists, a hepatologist, orthopedic surgeons, and gynecologists. Family practice residents rotate through the clinic and are taught the basic components of follow-up and potential health problems of childhood cancer survivors. The team also is active in speaking before oncologists and primary care physicians at continuing education programs.
All young adult survivors (n = 96) seen in the ACE Young Adult Program prior to January 31, 1999 form the study population. As noted in other settings, the percentage of survivors involved in long term follow-up decreases with age. Survivors seen in this program represent approximately 25% of the childhood cancer survivors who are at least age 18 years. To our knowledge the exact number and demographics of all adult survivors of childhood cancer who were treated at the study institution are not well described prior to the development of the cancer registry database in 1990, thus preventing comparison of the survivors followed in the ACE Young Adult Program with those who have been lost to follow-up. The current age of the 96 patients ranged from 18–34 years, with a median of 22.8 years. Patients were diagnosed with childhood cancer between 1973 and 1992, at a median age of 9.8 years (range, 1–18 years). The median interval since diagnosis was 15.2 years (range, 6–25 years). Approximately 59% of the patients were males and 41% were females. Of the 96 adult survivors, 86% were white, 8% were Hispanic, 5% were African-American, and 1% was classified as other.
As illustrated in Table 2, the primary cancers of the 96 young adults were acute lymphoblastic leukemia (ALL) (29%), Hodgkin disease (15%), non-Hodgkin lymphoma (12%), osteosarcoma (9%), Wilms' tumor (9%), and rhabdomyosarcoma (6%). Survivors of central nervous system (CNS) tumors were underrepresented in this population because they are followed separately by a multidisciplinary neurooncology team. Treatment modalities for the primary cancer included: chemotherapy only (22%); radiotherapy only (2%); chemotherapy and radiotherapy (32%); chemotherapy and surgery (17%); radiotherapy and surgery (5%); and chemotherapy, radiotherapy, and surgery (22%).
Table 2. Primary Cancers in 96 Consecutive Patients Seen in the ACE Young Adult Program
| Acute lymphoblastic leukemia||28|
| Acute myelogenous leukemia||4|
| Other sarcomas||8|
| Langerhans' cell histiocytosis||2|
| Hepatocellular carcinoma||1|
Grading of Late Effects
At the time of the current study, there were no universally accepted, validated, scoring systems available for use in grading late effects resulting from all modalities of treatment. Systems have been created for specific institutional12 or organizational13 use, but to our knowledge have yet to be validated. The late effects normal tissue (LENT) scoring system has been developed to grade radiation-related late effects but does not include scoring mechanisms for late effects secondary to surgery or chemotherapy.14–16 The Common Toxicity Criteria (CTC) was created in 1988 by the National Cancer Institute (NCI) as an additional outcome measure to compare the acute toxicities of different treatments. In 1998, the NCI released the second version of the CTC (CTCv2), which included the Radiation Therapy Oncology Group (RTOG)/European Organization for Research and Treatment of Cancer (EORTC) Late Radiation Morbidity Scoring Scheme.17 The CTCv2 guide is a familiar, widely used tool that grades toxicity from 0 to 4. Although CTCv2 was not developed specifically for use in grading late effects, the majority of late effects can be scored easily for the various organ systems or the appendixed Late Radiation Morbidity Scoring Scheme. For these reasons, the CTCv2 scoring system was selected in the current study to grade late effects. In Table 3, CTCv2 grading criteria of two late effects, left ventricular dysfunction and hypothyroidism, are presented as examples.
Table 3. Examples of Late Effects Graded by Common Toxicity Criteria v2
|Cardiac left ventricular function||Normal||Asymptomatic decline of resting EF of ≥ 10% but < 20% of baseline value; SF ≥ 24% but < 30%||Asymptomatic but resting EF below LLN for laboratory or decline of resting EF ≥ 20% of baseline value; < 24% SF||CHF responsive to treatment||Severe or refractory CHF requiring intubation|
|Hypothyroidism||Absent||Asymptomatic, TSH elevated, no therapy given||Symptomatic or thyroid replacement treatment given||Patient hospitalized for manifestations of hypothyroidism||Myxedema coma|
At each patient visit in the ACE Young Adult Program, all late effects diagnosed by means of history, physical examination, or testing are recorded in the database. Consecutive patients seen in the program between March 1, 1995 and January 31, 1999 were included in the study. Retrospectively, the database and each patient's medical chart were reviewed independently by two of the investigators (K.O. and D.E.) to grade the late effects recorded at the patient's last visit using the CTCv2. No discrepancies were found between the medical record and the database or between retrospective toxicity scoring and objective scoring with subsequent visits.
Subjective reports of symptoms that might overlap with other unrelated conditions, such as fatigue, dyspepsia, and insomnia, were not included. The only subjective symptom used for coding was pain due to radiation, and this was graded as per the CTCv2 guidelines. Weight gain, listed as a toxicity in the constitutional symptoms category, was used only for patients who had received cranial irradiation and who had documented weight gain within the guidelines of CTCv2. If a late effect secondary to radiation therapy could not be graded in a specific organ system, it was graded by the RTOG/EORTC Late Radiation Morbidity Scoring Scheme. For specific late effects not listed in the appropriate organ system category or in the Late Radiation Morbidity Scoring Scheme, the “organ system-other” category was used and the grade was assigned as: 0: none; 1: mild; 2: moderate; 3: severe; and 4: life-threatening. For two common groups of late effects, “Musculoskeletal” and “Infection,” the “Other” category was modified to a more detailed format, as shown in Table 4.
Table 4. Modified Categories of Common Toxicity Criteria v2
|Facial/skeletal abnormality||Absent||Mild facial or skeletal abnormality not requiring surgery and without chronic symptoms||Moderate facial or skeletal abnormality not requiring surgery but with chronic symptoms||An abnormality that is symptomatic and interfering with activities of daily living or requiring multiple plastic or reconstructive surgeries||Life-threatening or disabling problem|
|Hepatitis C||Absent||Normal ALT||Persistently elevated ALT (≥ 1.5 × normal)||Requiring treatment (interferon with or without ribavirin)||With liver failure|
Data were analyzed by descriptive techniques using frequencies, percentages, means, and medians as appropriate. The Student t test was used for comparison of survivors exposed to anthracyclines. Because of the small population studied, differences between cancer types were not analyzed. In addition, because the numbers of patients seen in the first 2 years of the program were small, comparisons over time were not calculated.
Approximately 31% (30 of 96) of the patients did not have a late effect measurable by the CTCv2 (Table 5). Thirty percent (9 of 30) of patients without a CTC-graded late effect had an “unclassified” late effect, including postnephrectomy (n = 6), asplenia (n = 1), depression (n = 1), and breast fibroadenoma in the radiation field ( n = 1).
Table 5. Late Effects Graded as per CTCV2
|No CTCv2 late effecta||31%||30/96|
| Single late effect||33%||32/96|
| Multiple late effects (≥ 2)||35%||34/96|
|Grade 3 or 4 toxicity||30%||29/96|
The 66 patients (69%) with a CTCv2-graded late effect had a total of 115 late effects. Approximately 33% of patients (32 of 96) had a single late effect whereas 35% had ≥ 2 late effects. Twenty patients (24%) were diagnosed with an asymptomatic late effect of Grade 2–4 (e.g., cardiomyopathy or hepatitis C) that previously was undetected and that subsequently required therapy or closer surveillance. Late effects diagnosed in the 66 patients are listed by involved organ system in Table 6.
Table 6. CTCv2 Grades of Late Effects By Organ System
|Coded only in late radiation|
| morbidity scoring||2||0||0||2|
|Total no. of late effects||115||42||36||37|
Twenty-nine of the survivors (30%) had a Grade 3 or 4 late effect that could be separated into life-threatening toxicities or sequelae predisposing to long term emotional distress. The CTCv2 categories of the ten patients who had potentially life-threatening toxicities were: Secondary Malignancy (synovial sarcoma and late [10 years] recurrence of Hodgkin disease); Cardiovascular (cardiomyopathy with congestive heart failure); Infection (hepatitis C with chronic active hepatitis requiring treatment or with liver failure); and Neurology (chronic seizure disorder). The other 19 patients had Grade 3 or 4 late effects that could be considered as nonlife-threatening sequelae that have a significant long term physiologic or psychologic burden. Categories of late effects included in this group were: Auditory (bilateral hearing loss requiring hearing aid); Cardiovascular (hypertension); Musculoskeletal (above-the-knee amputation, limb salvage, severe facial deformity after irradiation requiring plastic surgery, severe scoliosis with Harrington rod placement, and chronic pain); Neurology (profound cognitive dysfunction); Other Late Radiation Morbidity (recurrent small bowel obstruction, vaginal stenosis); and Sexual/Reproductive Function (infertility, ovarian failure, testicular atrophy).
Of the 27 patients who were exposed to a cumulative dose of an anthracycline ≥ 300 mg/m2, 13 patients (48%) had evidence of altered left ventricular function. Six survivors had Grade 1 changes (see Table 3 for grading criteria) whereas seven had Grade 2 changes. The mean cumulative anthracycline dose, age at treatment, and interval since treatment was not statistically different for survivors with and without left ventricular dysfunction. An additional three patients were diagnosed with Grade 2 left ventricular dysfunction who received mantle radiation only (two patients) or mantle radiation in combination with an anthracycline (one patient).
Eighteen patients were hepatitis C antibody positive, including 6 patients with persistently (≥ 6 months) elevated alanine aminotransferase levels ≥ 1.5 times normal, 3 patients with evidence of early cirrhosis who were treated with interferon-α, and 1 patient with advanced cirrhosis.
As childhood cancer survivors enter adulthood, new questions arise regarding their long term health care. Should the adult survivor be followed in a pediatric institution? What role will primary care physicians play in the health care of this population? Rosen emphasizes that young adult survivors need health care provided by adult-oriented physicians.8 Although transition programs have been suggested as a potential method of long term follow-up of young adult survivors, to our knowledge little has been written regarding the structure and health care provided by such a program.7–11 The current study presents an example of an interdisciplinary transition program based in an adult-oriented ambulatory care setting that links the expertise of the pediatric oncology team, family physicians, and a network of pediatric and adult subspecialists.
Late effects commonly are diagnosed in young adult survivors of childhood cancer who are followed longitudinally. Approximately 69% of the patients reported in the current study had at least one late effect of their cancer therapy. Late effects varied in severity, with nearly one-third of all patients having moderate to severe late effects. Musculoskeletal abnormalities, left ventricular dysfunction, hepatitis C, and hypothyroidism were the most common late effects seen in this population of young adult survivors. The prevalence of left ventricular dysfunction seen in this population was similar to that reported in other follow-up studies.18–20 Similarly, 32% of the ALL survivors followed in the ACE Young Adult Program were positive for the hepatitis C virus, which is within the 23–49% range of prevalence reported in longitudinal studies.21–23
The findings of the current study are similar to those noted in populations of childhood cancer survivors seen in long term follow-up clinics in three other studies (Table 7).12, 13, 24 With the exception of CNS tumors, the cancer groups seen in the patient population in the current study are similar to those reported in these three studies. The percent of childhood cancer survivors with a late effect diagnosed in long term follow-up has been reported to range from 58–71% in these studies. Multiple late effects are common, occurring in approximately 32–49% of patients. The Swiss Pediatric Oncology Group (SPOG) study reported that 43% (13 of 30) of their patients were diagnosed with a late effect that previously was undetected and required therapy or closer surveillance.13 Similarly, we reported that 24% of patients in the current study had an asymptomatic late effect of CTCv2 Grade 2–4 (identified by us for the first time) that necessitated treatment or close follow-up and/or further evaluation. The data from the current study also add to the findings of Garré et al.12 that a significant portion of childhood cancer survivors have moderate to severe late effects that require long term treatment, affect their daily activities, or require continued close surveillance.
Table 7. Late Effects Diagnosed in Adolescent and Young Adult Survivors of Childhood Cancer in Four Single Institution Studies
|No. of patients||288||30||290||96|
|Setting||Children's Hospital||Children's Hospital||Children's Hospital||Adult setting|
|Median age, current age (yrs)||15||15||23||23|
|Median interval (yrs)||8||NR||14||15|
|Late effects, % of patients||69%||67%||58%||69%|
|≥2 late effects, % of patients||49%||NR||32%||36%|
|Grading system||Institution specific||SPOG||None used||CTCv2|
|Late effect Grade ≥3, % of patients||25%||NR||NR||30%|
|Primary cancer types|
| Hodgkin disease||15%||10%||15%||15%|
| Wilms' tumor||13%||10%||11%||9%|
| CNS tumor||6%||4%||11%||1%|
Grading of late effects can provide useful information for comparing treatment and long term follow-up strategies. The current report adds to the findings of two other studies in the use of a grading system for late effects.12, 13 To our knowledge, as yet there is no universally accepted grading system that has been developed and validated solely for the study of late effects related to all treatment modalities. Garré et al. developed a set of criteria to grade late effects, with toxicity grades including: Grade 0: no late effect; Grade 1: asymptomatic changes that did not require any corrective measure and did not influence the subject's general activity; Grade 2: moderate symptomatic changes interfering with the subject's activity; Grade 3: severe symptomatic changes that required major corrective measures and strict and prolonged surveillance; and Grade 4: life-threatening sequelae.12 SPOG developed a grading system for late effects in childhood cancer survivors and piloted its use in 30 patients, half of whom were age > 16 years.13 However, to our knowledge validation studies of the SPOG grading system have not been published. The SPOG system also ranges from 0 to 4, with Grade 0: no late effect; Grade 1: asymptomatic patient requiring no therapy; Grade 2: asymptomatic patient, requires continuous therapy or continuous medical follow-up or symptomatic late effects resulting in reduced school, job, or psychosocial adjustment while remaining fully independent; Grade 3: physical or mental sequelae not likely to be improved by therapy but able to work at least partially; and Grade 4: severely handicapped patients unable to work independently.
In general, the CTCv2 guidelines are similar to the grading systems described earlier. However, the CTCv2 grading scheme is more detailed, including criteria specific for the majority of organ systems affected. These organ specific criteria should reduce the interrater variability, although further validation in the long term cancer survivor population is needed to render it more applicable to the late effects setting. The only two categories in which the “Other” category was repeatedly used in our study were “Musculoskeletal” and “Infection,” and the grading scheme was modified as discussed earlier. Otherwise, late effects were graded easily with the CTCv2.
An additional finding that is important to note is the underrepresentation of minorities who have been seen in the ACE Young Adult Program. African-American and Hispanic patients comprised 33% of the children newly diagnosed with cancer at the Children's Medical Center of Dallas between 1993–1997, whereas thus far they have comprised only 14% of the patients seen in the ACE Young Adult Program. This is similar to the findings of the National Wilms' Tumor Study Group, which reported a 10-year cumulative loss to follow-up of 45.5% for African-American survivors and 36.5% for Hispanic survivors compared with 21.5% for white, non-Hispanic survivors.25 To understand better the potential barriers to long term follow-up faced by underrepresented or underserved minorities, the investigators of the ACE Young Adult Program currently are analyzing data of a retrospective cohort of childhood cancer patients diagnosed between 1990–1995 at Children's Medical Center of Dallas. Socioeconomic, cancer-related, and gender-/age-related factors that might predict loss to follow-up are being evaluated.
Several limitations should be recognized in the interpretation of the current study. The sample size was relatively small, limiting analysis between cancer types or with longitudinal follow-up. Second, the CTC grades were assigned by means of a retrospective review of the diagnosed late effects. For reasonably valid comparisons of treatment groups and cancers, a validated grading system used prospectively will be essential. Third, the current study population lacked patients with CNS tumors. Finally, a selection bias is introduced by the very nature of a long term follow-up program. Only survivors who decided to come for evaluation were available for assessment and comparison. This may overrepresent patients with symptomatic late effects who are more likely to be concerned about their health. However, it should be noted that a substantial number of patients (24%) who were evaluated were asymptomatic and were diagnosed with a moderate to severe late effect that was unrecognized previously.
There is still much that is not understood regarding the health problems of and the delivery of care to adult survivors of childhood cancer. Research to date has focused on the late effects during the first 10–15 years after therapy and now it is necessary to expand this body of knowledge to the adult years when many new variables arise. What effect will life-style and risk behavior, such as smoking, have on the incidence rate of second primary malignancies? How will previous therapies, such as mantle radiation, affect the progression of common adult health problems such as coronary artery disease? Perhaps equally important are outcome studies evaluating the cost-effectiveness of different screening and surveillance strategies and secondary prevention interventions. As survivors move into the adult years, start their careers and families, and move away from their parents and their treating institutions, more and more will seek their health care from primary care physicians. How will the communication between pediatric cancer centers and primary care physicians be facilitated? Finally, as we enter this new millennium it is essential that this growing population of aging childhood cancer survivors be viewed as a high risk, underserved group that has potential for widely diverse physiologic and psychosocial problems and faces significant impediments to the access of medical care in our current health care system.
The current study describes a multidisciplinary transition program in an ambulatory, adult-based setting for young adult survivors of childhood cancer, a program that might be viewed as a model for others. These findings add to the growing knowledge base that supports the necessity of long term, longitudinal follow-up of childhood cancer survivors who have reached adulthood and demonstrate the potential value of grading of late effects.