The cancer screening practices of adult survivors of childhood cancer

A report from the Childhood Cancer Survivor Study

Authors

  • Mark W. Yeazel M.D., M.P.H.,

    Corresponding author
    1. Department of Family Practice and Community Health, University of Minnesota Medical School, Minneapolis, Minnesota
    • Department of Family Practice and Community Health, University of Minnesota Medical School, Mayo Medical Code 381, 420 Delaware Street S.E., Minneapolis, MN 55455
    Search for more papers by this author
    • Fax: (612) 625-0640

  • Kevin C. Oeffinger M.D.,

    1. Department of Family Practice and Community Medicine, University of Texas, Southwestern Medical Center at Dallas, Texas
    Search for more papers by this author
  • James G. Gurney Ph.D.,

    1. Department of Pediatrics, University of Minnesota Medical School, Minneapolis, Minnesota
    Search for more papers by this author
  • Ann C. Mertens Ph.D.,

    1. Department of Pediatrics, University of Minnesota Medical School, Minneapolis, Minnesota
    Search for more papers by this author
  • Melissa M. Hudson M.D.,

    1. Department of Hematology Oncology, St. Jude Children's Research Hospital and University of Tennessee College of Medicine, Memphis, Tennessee
    Search for more papers by this author
  • Karen M. Emmons Ph.D.,

    1. Division of Community-Based Research, Dana Farber Cancer Institute and Harvard School of Public Health, Boston, Massachusetts
    Search for more papers by this author
  • Hegang Chen Ph.D.,

    1. Department of Biostatics and Bioinformatics, University of Maryland School of Medicine, College Park, Maryland
    Search for more papers by this author
  • Leslie L. Robison Ph.D.

    1. Department of Pediatrics, University of Minnesota Medical School, Minneapolis, Minnesota
    Search for more papers by this author

  • The following are Childhood Cancer Survivor Study (CCSS) Institutions and Investigators: University of California, San Francisco, CA (Clinical center contributing cases) (Arthur Ablin, M.D. [Institutional Principal Investigator]); University of Alabama, Birmingham, AL (Clinical center contributing cases) (Roger Berkow, M.D. [Institutional Principal Investigator]); International Epidemiology Institute, Rockville, MD (John D. Boice, Jr., Sc.D. [Member CCSS Steering Committee]); University of Washington, Seattle, WA (Norman Breslow, Ph.D. [Institutional Principal Investigator]); University of Texas-Southwestern Medical Center at Dallas, TX (Clinical center contributing cases) (George R. Buchanan, M.D. [Institutional Principal Investigator]); Dana-Farber Cancer Institute, Boston, MA (Clinical center contributing cases) (Lisa Diller, M.D. [Institutional Principal Investigator], Holcombe Grier, M.D. [Former Institutional Principal Investigator], and Frederick Li, M.D. [Member CCSS Steering Committee]); Texas Children's Center, Houston, TX (Clinical center contributing cases) (Zoann Dreyer, M.D. [Institutional Principal Investigator]); Children's Hospital and Medical Center, Seattle, WA (Clinical center contributing cases) (Debra Friedman, M.D., M.P.H. [Institutional Principal Investigator] and Thomas Pendergrass, M.D. [Former Institutional Principal Investigator]); Roswell Park Cancer Institute, Buffalo, NY (Clinical center contributing cases) (Daniel M. Green, M.D. [Institutional Principal Investigator and Member of the CCSS Steering Committee]); Hospital for Sick Children, Toronto, ON (Clinical center contributing cases) (Mark Greenberg, M.B., Ch.B. [Institutional Principal Investigator]); St. Louis Children's Hospital, St. Louis, MO (Clinical center contributing cases) (Robert Hayasyhi, M.D. [Institutional Principal Investigator] and Teresa Vietti, M.D. [Former Institutional Principal Investigator]); St. Jude Children's Research Hospital, Memphis, TN (Clinical center contributing cases) (Melissa Hudson, M.D. [Institutional Principal Investigator and Member of the CCSS Steering Committee]); University of Michigan, Ann Arbor, MI (Clinical center contributing cases) Raymond Hutchinson, M.D. [Institutional Principal Investigator]); Stanford University School of Medicine, Stanford, CA (Clinical center contributing cases) (Michael P. Link, M.D. [Institutional Principal Investigator] and Sarah S. Donaldson, M.D. [Member of the CCSS Steering Committee]; Children's Hospital of Philadelphia, PA (Clinical center contributing cases) (Anna Meadows, M.D. [Institutional Principal Investigator and Member of the CCSS Steering Committee] and Bobbie Bayton [Member of the CCSS Steering Committee]); Children's Hospital, Oklahoma City, OK (John Mulvihill, M.D. [Member of the CCSS Steering Committee]); Children's Hospital, Denver, CO (Clinical center contributing cases) (Brian Greffe, M.D. [Institutional Principal Investigator] and Lorrie Odom, M.D. [Former Institutional Principal Investigator]); Children's Health Care-Minneapolis, MN (Clinical center contributing cases) (Maura O'Leary, M.D. [Institutional Principal Investigator]); Columbus Children's Hospital, OH (Clinical center contributing cases) (Amanda Termuhlen, M.D. [Institutional Principal Investigator], Frederick Ruymann, M.D. [Former Institutional Principal Investigator], and Stephen Qualman, M.D. [Member of the CCSS Steering Committee]); Children's National Medical Center, Washington, D.C. (Clinical center contributing cases) (Gregory Reaman, M.D. [Institutional Principal Investigator] and Roger Packer, M.D. [Member of the CCSS Steering Committee]); Children's Hospital of Pittsburgh, PA (Clinical center contributing cases) (A. Kim Ritchey, M.D. [Institutional Principal Investigator] and Julie Blatt, M.D. [Member of the CCSS Steering Committee]); University of Minnesota, Minneapolis, MN (Clinical center contributing cases) (Leslie L. Robison, Ph.D. [Institutional Principal Investigator and Member of the CCSS Steering Committee], Ann Mertens, Ph.D. [Member of the CCSS Steering Committee], Joseph Neglia, M.D., M.P.H. [Former Institutional Principal Investigator], Mark Nesbit, M.D. [Member of the CCSS Steering Committee], and Stella Davies, M.D., Ph.D. [Member of the CCSS Steering Committee]); Children's Hospital of Los Angeles, CA (Clinical center contributing cases) (Kathy Ruccione, R.N., M.P.H. [Institutional Principal Investigator]); Memorial Sloan Kettering Cancer Center, New York (Clinical center contributing cases) (Charles Sklar, M.D. [Institutional Principal Investigator and Member of the CCSS Steering Committee]); National Cancer Institute, Bethesda, MD (Malcolm Smith, M.D. [Member of the CCSS Steering Committee] and Martha Linet, M.D. [Member of the CCSS Steering Committee]); Mayo Clinic, Rochester, MN (Clinical center contributing cases) (W. Anthony Smithson, M.D. [Institutional Principal Investigator] and Gerald Gilchrist, M.D. [Former Institutional Principal Investigator]); University of Texas M. D. Anderson Cancer Center, Houston, TX (Clinical center contributing cases) Louise Strong, M.D. [Institutional Principal Investigator] and Marilyn Stovall, Ph.D. [Member of the CCSS Steering Committee]); Riley Hospital for Children, Indianapolis, IN (Clinical center contributing cases) Terry A. Vik, M.D. [Institutional Principal Investigator] and Robert Weetman, M.D. [Former Institutional Principal Investigator]); Fred Hutchinson Cancer Center, Seattle, WA (Yukata Yasui, Ph.D. [Institutional Principal Investigator] and John Potter, M.D., Ph.D. [Former Institutional Principal Investigator and Member of the CCSS Steering Committee]); and University of California-Los Angeles, CA (Including Miller Children's Hospital, Long Beach and Children's Hospital of Orange County) (Clinical center contributing cases) (Lonnie Zeltzer, M.D. [Institutional Principal Investigator and Member of the CCSS Steering Committee]).

Abstract

BACKGROUND

The current study characterized the self-reported cancer screening practices of adult survivors of childhood cancer.

METHODS

A cohort of 9434 long-term survivors of childhood cancer and a comparison group of 2667 siblings completed a 289-item survey that included items regarding cancer-screening practices.

RESULTS

Overall, 27.3% of female respondents reported performing breast self-examination (BSE) regularly, 78.2% reported undergoing a Papanicolaou smear within the previous 3 years, 62.4% underwent a clinical breast examination (CBE) within the last year, and 20.9% had gotten a mammogram at least once in their lifetime. Approximately 17.4% of male respondents reported performing regular testicular self-examination (TSE). Women age ≥ 30 years who had been exposed to chest or mantle radiation therapy were more likely to report undergoing CBE (odds ratio [OR], 1.59; 95% confidence interval [95% CI], 1.32–1.92) and mammography (OR, 1.92; 95% CI, 1.47–2.56). Compared with the sibling comparison group, survivors demonstrated an increased likelihood of performing TSE (OR, 1.52; 95% CI, 1.22–1.85) or BSE (OR, 1.30; 95% CI, 1.10–1.52), of having undergone a CBE within the last year (OR, 1.18; 95% CI, 1.02–1.35), and of ever having undergone a mammogram (OR, 1.82; 95% CI, 1.52–2.17).

CONCLUSIONS

The results of the current study demonstrate that the cancer screening practices among survivors of childhood cancer are below optimal levels. Primary care physicians who include childhood cancer survivors among their patients could benefit these individuals by informing them about future cancer risks and recommending appropriate evidence-based screening. Cancer 2004. © 2003 American Cancer Society.

Largely because of improvements in cancer therapy over the last several decades, more children are surviving cancer than ever before. Since 1992, the overall 5-year survival rate for children diagnosed with cancer has been reported to be 77%.1 With these improved survival outcomes, there also is an increasing prevalence of survivors who are at risk for a variety of long-term late effects of cancer and its treatment. A study of 5-year survivors of childhood cancer found an overall standardized mortality ratio that was 10.8 times greater than that of the general U.S. population.2 Recurrence of the original malignancy is a major contributor, although mortality risk is also increased because of subsequent malignancies, cardiomyopathy and ischemic heart disease, and pulmonary fibrosis.2 Long-term survivors also may suffer morbidity because of late effects of their cancer therapy. Several small studies have demonstrated that approximately 2 in 3 long-term survivors will be affected by at least 1 late effect of therapy, with approximately 30% developing moderate to severe effects.3–6 These late effects potentially include secondary neoplasms; pulmonary effects; cardiovascular effects including anthracycline-induced cardiomyopathy, ischemic heart disease, and obesity; osteopenia and osteoporosis; endocrine disorders including growth hormone deficiency and thyroid disease; cognitive dysfunction; and dental and periodontal disease.7–15

The potential of being afflicted by these late health effects emphasizes the need for preventive health practices among childhood cancer survivors. Because recurrent or secondary tumors will have such a profound impact on a survivor's health, cancer screening is of particular importance.

Surprisingly, to our knowledge there are very few studies published to date describing the cancer screening practices of childhood cancer survivors. In a randomized controlled trial evaluating a multicomponent behavioral intervention to promote health protective behaviors in a group of 272 survivors of childhood cancer, Hudson et al. observed that at baseline, 27% of the subjects reported performing monthly breast self-examination (BSE) or testicular self-examination (TSE).16 In another trial evaluating the breast cancer screening practices of 90 adolescent and young adult female survivors of Hodgkin disease believed to be at high-risk of developing secondary breast cancer, only 9% reported performing regular BSE.17 Although 72% of that cohort had undergone mammography screening, only 47% had undergone a mammogram within the previous 2 years. These studies suggest that survivors generally report inadequate cancer screening practices despite their increased vulnerability to second malignancies.

The Childhood Cancer Survivor Study offers a unique opportunity to characterize the cancer screening practices of adult survivors of childhood cancer. Specifically, we sought to evaluate the following questions: 1) what are the current self-reported cancer screening practices of this large cohort of young adult survivors of childhood cancer? Do these practices differ by sociodemographic factors or by cancer-related factors such as cancer diagnosis, age at diagnosis, or interval from cancer diagnosis? 2) do screening practices differ from a randomly selected comparison group of siblings without cancer? 3) are the breast cancer screening practices of women exposed to chest or mantle radiation therapy, a particularly high-risk group, different from those of other childhood cancer survivors?

MATERIALS AND METHODS

This analysis is part of the Childhood Cancer Survivor Study, a 25-site multiinstitutional study established in 1994 that follows a cohort of individuals diagnosed with cancer, leukemia, tumor, or other similar illnesses during childhood. Eligibility criteria include: 1) diagnosis and initial treatment of leukemia, central nervous system tumor (excluding meningioma and craniopharyngioma), Hodgkin disease, non-Hodgkin lymphoma, neuroblastoma, soft tissue sarcoma, kidney cancer, or bone cancer; 2) a diagnosis date between January 1, 1970 and December 31, 1986; 3) age < 21 years at the time of diagnosis; 4) survival of > 5 years after diagnosis; and 5) English speaking.

The Childhood Cancer Survivor Study protocol was reviewed and approved by the Human Subjects Committees at each participating institution, and informed consent was obtained for all participants. After initial contact by the treating institution, participants were contacted requesting their participation in the study. Beginning August 1, 1994, participants completed a 289-item questionnaire assessing medical late effects, current medical conditions, medication use, health status, health behaviors, pregnancy history, demographic characteristics, socioeconomic indicators, insurance coverage, and other information. Although data collection continued for several years, the majority of surveys were returned in 1995 and 1996. In addition, trained data abstractors at the treating institutions reviewed medical records to confirm cancer diagnoses and record treatment information. Copies of the questionnaire and the medical record abstraction form are available from URL: www.cancer.umn.edu/ccss.

A total of 20,276 eligible patients were identified. Of these, 2996 patients (15%) were lost to follow-up, 3132 (15%) refused participation, and 94 (< 1%) were pending return of a baseline questionnaire at the time of last follow-up. A total of 14,054 survivors participated, which represents 69% of those potentially eligible or 81% of those successfully contacted. Medical record release of information was obtained from 12,752 subjects and medical record abstraction was completed for 12,455 subjects. The current report includes the 9434 of these participants who were alive and at least 18 years old at the time of interview.

A sibling cohort was assembled by identifying 5857 siblings of cancer survivors who were randomly selected for study participation. Of these selected siblings, 5800 (99%) had been contacted successfully and 3585 (61%) had completed the questionnaire at the time of last follow-up. From the sibling cohort, 2667 individuals were age ≥ 18 years at the time of the interview and were used as a comparison group in the current analysis. The study design and cohort characteristics have been described in further detail elsewhere.18

Cancer Screening Outcome Measures

Females were asked the following questions: 1) “How often do you perform monthly breast self-examination?” with responses of “regularly (once a month),” “occasionally,” and “rarely or never”; 2) “when was the last time you had a breast examination by a doctor or health care professional?” with responses recoded as a dichotomous variable of < 1 year ago compared with > 1 year ago; 3) “have you ever had a mammogram?” with responses of no or yes; and 4) “when was the last time you had a [Papanicolaou] pap smear (test for cancer of the cervix) ?” with responses recoded as a dichotomous variable of within the last 3 years compared with ≥ 3 years ago. Males were asked about TSE with the following question: “How often do you perform monthly testicular self-examination?” with choices of “regularly (once a month),” “occasionally,” and “rarely or never.”

The rationale for this coding scheme was consistent with the most commonly acknowledged sources of recommendations for screenings of the general population at the time this survey was conducted, including those of the U.S. Preventive Services Task Force (USPSTF) and the American Cancer Society (ACS). The guidelines for the general population were chosen because to our knowledge there is no consensus concerning guidelines for cancer screening in childhood cancer survivors. These general population guidelines may underestimate what actually is needed for cancer survivors. Using the general population guidelines, cervical cancer screening by Papanicolaou (Pap) smear was recommended for every woman age ≥ 18 years at a 1–3-year interval.19–21 The decision whether to screen less frequently than annually was to be based on the presence or absence of risk factors for cervical cancer. Yearly clinical breast examination (CBE) for those women age < 40 years was recommended by the American College of Obstetricians and Gynecologists (ACOG),19 whereas the American Academy of Family Practice (AAFP) recommended that examinations be performed on an 1–3-year basis.22 Because of the increased risk of breast cancer for many survivors, we chose to use CBE within the last year as our outcome measure. Both the ACS and ACOG recommend that women exam their breasts monthly.19, 20 The ACS, AAFP, and the American Urological Association encouraged regular TSE.22–24 Because we sought to describe survivors' current self-examination practices, we differentiated between those who perform self-examination regularly and those who occasionally or never performed this examination.

Independent Variables

Information was obtained regarding demographic and socioeconomic variables such as gender, age at interview, ethnicity, highest educational attainment, household income, and health insurance coverage. Ethnicity was a categoric variable defined as white and non-Hispanic, or minority. Information also was obtained concerning cancer-related variables including the type of cancer, the age at diagnosis, and the time interval from cancer diagnosis to interview. Subjects also were asked to rate their general health status with the question, “Would you say that your health is…” Responses were poor, fair, good, very good, or excellent. Subjects also were asked to rate how concerned they were about their future health on a five-point Likert scale ranging from very concerned to not at all concerned.

Data Analysis

Comparisons within strata of independent variables were assessed using the Pearson chi-square test with two-tailed tests for significance. Multivariate risk factors predicting the cancer screening practices of survivors were calculated using logistic regression. Adjusted odds ratios (ORs) for comparisons of survivors and the sibling control group were calculated using multivariate logistic regression models adjusting for age, ethnicity, education, income, and health insurance status. To account for potential within-family correlation between the survivor and his/her sibling from the same family, variances were calculated using appropriate generalized estimating equations for correlated data. For multivariate analyses of the survivors' mammography screening practices, we chose to stratify subjects on the basis of age ≥ 30 years, because age 30 years represents a conservative age at which many authorities agree mammography should be performed regularly for cancer survivors who were treated with chest or mantle radiation therapy.25, 26

RESULTS

Cancer Screening Practices of Survivors

Table 1 describes the cancer screening practices of adult survivors of childhood cancer with regard to cancer-related factors. Chi-square analyses were used to test for differences in reported cancer screening practices between strata of each cancer-related factor. Several differences were demonstrated. For example, survivors who were older at the time of the original cancer diagnosis reported higher rates of each cancer screening practice compared with those survivors who were diagnosed at younger ages. Subjects who perceived their health as excellent reported the highest rate of CBE and the lowest rates of mammography, whereas those whose perceived health was poor reported the lowest rates of CBE and the highest rates of mammography. In general, subjects with higher levels of future health concerns reported higher levels of performance of all cancer screening behaviors.

Table 1. The Cancer Screening Practices of Adult Survivors of Childhood Cancer: Cancer-Related Factors
inline image

Multivariate analyses describing the cancer screening practices of survivors with regard to sociodemographic factors are shown in Table 2. The analyses controlled for age, ethnicity, education, health insurance status, concern for future health, and chest or mantle radiation therapy. Female survivors age ≥ 25 years were more likely than those ages 18–24 years to report undergoing Pap testing within 3 years or performing BSE regularly. Survivors who were college graduates were more likely to report a recent CBE or Pap test compared with those with lower levels of educational attainment, as were women with health insurance compared with women without insurance. Survivors with lower levels of future health concern were less likely to report all the cancer screening practices compared with those with higher levels of concern.

Table 2. Multivariate Risk Factors Predicting the Cancer Screening Practices of Adult Survivors of Childhood Cancer
inline image

Breast Cancer Screening by Survivors in a High-Risk Group

Overall, 34.0% of survivors age ≥ 30 years who received chest or mantle radiation therapy performed BSE compared with 28.8% of survivors without chest or mantle radiation therapy (P = 0.05) (data not shown). In addition, 73.4% of survivors age ≥ 30 years who received chest or mantle radiation therapy reported a recent CBE compared with 64.3% of survivors without chest or mantle radiation therapy (P < 0.001) (data not shown) and 56.9% of survivors age ≥ 30 with chest or mantle radiation therapy reported ever having a mammogram, compared with 35.5% of survivors without chest or mantle radiation therapy (P < 0.001) (data not shown).

The analyses in Table 2 suggest that female survivors age ≥ 30 years who were exposed to chest or mantle radiation therapy were more likely to report undergoing a recent CBE (OR, 1.59; 95% confidence interval [95% CI], 1.31–1.92) or to ever have had a mammogram (OR, 1.92; 95% CI, 1.47–2.56) compared with other female cancer survivors. No substantive difference was noted between these groups with regard to the likelihood of performing a regular BSE.

Survivor-Sibling Comparisons

Comparisons of the cancer screening practices of adult long-term survivors with a sibling comparison group are shown in Table 3. ORs for each cancer screening practice are adjusted for age, ethnicity, education, income and health insurance. With the exception of cervical cancer screening, survivors demonstrated a modestly increased likelihood of performing each cancer screening practice.

Table 3. ORs for Reporting Cancer Screening Practices in Adult Survivors of Childhood Cancer in Comparison with Siblings
VariableORa95% CIP value
  • OR: odds ratio; 95% CI: 95% confidence interval; Pap: Papanicolaou.

  • a

    Logistic regression analyses using generalized estimating equation models, adjusted for age, race, education, income, and health insurance.

Females   
 Breast self-examination, monthly1.301.10–1.520.0013
 Clinical breast examination within 1 year1.181.02–1.350.0270
 Mammogram, ever1.821.52–2.17< 0.0001
 Pap smear within last 3 years0.830.69–1.010.0582
Males   
 Testicular self-examination, monthly1.521.22–1.850.0002

DISCUSSION

The data from the current study provide evidence that cancer screening practices among young adult survivors of childhood cancer are below optimal levels of that often recommended for the general population. Only 78% of all female survivors reported herein had undergone a Pap test within the last 3 years. Less than 30% of female survivors reported practicing BSE, and only 62% reported undergoing a CBE within the last year. Approximately 21% of female survivors reported ever having had a mammogram, although many were not at an age at which screening is recommended in the general population. Only 17% of male survivors reported performing TSE.

Some interesting findings resulted from the multivariate analyses that describe risk factors for reporting cancer-screening practices. The lack of health insurance was associated with a decreased likelihood of undergoing a CBE or Pap test, both of which require medical care visits and recommendations of clinicians. No substantive differences were noted between insured and uninsured survivors with regard to mammography. A lack of health insurance was not associated with TSE or BSE, practices performed by the survivors themselves that do not require medical care visits. Survivors who expressed a lack of concern for future health issues were less likely to report cancer screening practices for all the practices described in the current study. This is consistent with constructs from the Health Belief Model of behavior change that suggest the perceived susceptibility and severity of a health condition are important motivators of behavior aimed at reducing the perceived threat.27 The youngest female survivors (ages 18–24 years) reported in the current study were less likely to report cancer screening practices compared with older survivors. Survivors of both genders with a low level of educational attainment were less likely to report cancer screening practices (except mammography). To the extent we could evaluate them, ethnicity or race were not found to be independently associated with screening practices. These analyses illustrate the characteristics of survivors least likely to perform cancer screening practices: those who are younger, have lower levels of educational attainment, lack concern for future health, and lack health insurance.

With the exception of cervical cancer screening, a modestly increased likelihood of screening was noted among survivors when compared with siblings for each of the practices we evaluated, after accounting for age, ethnicity, education, income, and health insurance. It should be noted that at the time of this study, screening mammography for women of average risk was recommended for those age ≥ 50 years, and the role of mammography for younger women (usually those age ≥ 40 years) was debated. Based on these guidelines, many of the study subjects may have undergone mammography for diagnostic evaluation instead of screening. One limitation of the current study was that we could not distinguish between screening and diagnostic mammograms.

Survivors who are at increased risk for breast cancer because of chest or mantle radiation therapy demonstrated below optimal breast cancer screening practices. Among those aged ≥ 30 years, only 56.9% reported ever having a mammogram, 34.0% reportedly practiced regular BSE, and 73.4% reported undergoing a CBE within the past year. Although these percentages were higher than those for survivors without previous chest or mantle radiation exposure, there still was a substantive proportion of survivors who were not practicing optimum risk-based screening. It should be recognized that much of the information regarding the risk of a second malignant neoplasm (including breast cancer) has been published since the mid-1990s, and many of the survivors may not have known that they were potentially in a high-risk group.

Admittedly, the efficacy of both TSE28, 29 and BSE in the general population recently has been questioned.30, 31 For example, although TSE is a highly sensitive screen, there reportedly is poor specificity and a low positive predictive value for this low-prevalence cancer.29 It has been estimated that 50,000 men in England and Wales would need to perform TSE monthly for 10 years to prevent 1 death.32 While recognizing the current controversy surrounding the use of these examinations in the general population, both TSE and BSE were advocated by many groups during the time period of the current study. Self-examination techniques also may play a role for childhood cancer survivors as a means of increasing awareness concerning physical changes after cancer.

The case for the importance of BSE in young women who have undergone chest radiation therapy is more compelling. There is strong consensus regarding the need for increased breast cancer surveillance in this group, although specific recommendations vary.17, 25, 26, 33–35 Although mammography is the standard screening modality for breast cancer, it is reported to be limited in its ability to evaluate dense premenopausal breast tissue and represents additional radiation exposure.36, 37 Recognizing the limitations of mammography, as well as the excess risk of breast cancer in this group of survivors, the Childrens Oncology Group Late Effects Study Group (COG-LESG) recently recommended a surveillance program including mammography, CBE, and BSE. BSE was recommended for this high-risk group despite its failure to reduce breast cancer mortality in the context of a randomized controlled trial that evaluated teaching women at low risk to perform the technique, because of its association with a shorter lag time from the appearance of clinical symptoms to the diagnosis of breast cancer, earlier clinical stage, and smaller tumor size.31, 38–40 The COG-LESG recommended yearly CBE beginning at puberty, with twice-yearly exams starting at age 25 years, annual mammography initiated 8 years after the completion of radiation therapy or age 25 years (whichever comes last), and monthly BSE.41

The current study has several limitations that should be considered when interpreting the results. These are self-reported data regarding cancer screening practices and the validity of these reports has, to our knowledge, not been determined. Volunteer bias is possible and participants may be healthier than nonparticipants, or may be more likely to practice preventive health behaviors. In this case, our estimates of the frequency of cancer screening practices of survivors would likely be an overestimate of their actual practices. It is possible that the cancer screening practices of the sibling comparison group differ systematically from the general population. However, our findings for the sibling group approximate those of a national sample of respondents to the 1995 Behavioral Risk Factor Surveillance System (BRFSS) survey. For example, the sibling group demonstrated fairly similar rates compared with the national sample of reporting ever having a mammogram among those ages 18–39 years (20% vs. 23%) and ages 40–49 years (80% vs. 81%)42 as well as reporting undergoing a Pap smear within the previous 3 years among those ages 18–49 years (85% vs. 88%).43 The majority of this information was collected between 1995 and 1996, and screening practices may have changed since that time. Lastly, our cohort had only limited ability to generalize our findings to minority survivors.

The study limitations notwithstanding, several important implications from this analysis can be drawn. Because of the risk of second malignant neoplasm or recurrence of their original cancer, survivors should at a minimum adhere to screening recommendations for the average risk population. Our belief is that some subgroups of survivors should be screened more intensively because of their high-risk status. For example, the increased risk of breast cancer among many survivors suggests that breast cancer screening methods should utilize at least mammography and CBE. However intuitive it appears, it has not been unequivocally determined whether cancer survivors represent a high-risk group that would benefit from these or other cancer screening practices. Again, this emphasizes the need to develop further the body of evidence that forms the basis for making screening recommendations for childhood cancer survivors.

Methods to improve the cancer screening behaviors of survivors are needed. Survivors of childhood cancer and their primary care physicians need information regarding their future cancer risks and appropriate screening guidelines. Effective communication of a survivor's medical history to a clinician also is essential to ensure appropriate screening. However, a recent study suggested that only 72% of survivors could accurately report their diagnosis with precision, and even fewer could accurately report on their cancer-associated therapies.44 Survivors need the means to provide clinicians with accurate information concerning their cancer diagnosis and treatment. It is incumbent on the physicians providing care for pediatric cancer patients to provide them with an accurate summary of their treatment history, and information concerning the potential late effects of those therapies. Mechanisms to insure adequate communication of a survivor's medical history between survivors and physicians are needed. In addition, interventions are needed to increase appropriate cancer-screening behaviors, targeting those who are at highest risk for developing cancer, such as those who have undergone chest or mantle radiation therapy. Efforts addressing the primary prevention of cancer, such as smoking cessation efforts, also are needed.

It also is important to focus interventions for improving survivors' cancer-screening practices toward their primary care physicians. Clinicians need to understand that cancer-screening recommendations for survivors of childhood cancer may differ from those recommended for the general population. There is a need to further develop the body of evidence that supports screening guidelines for survivors, and find effective methods with which to incorporate these guidelines into clinical practice. With interventions such as these, it should be possible to improve the cancer screening practices of childhood cancer survivors, ultimately reducing the morbidity and mortality associated with the recurrence of their original malignancy or second malignant neoplasms.

Ancillary