|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 31 (November 1), 2007: pp. 4929-4932 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.12.1905 Incidence of Neuroblastoma After a Screening Program
From the Hôpital Ste-Justine, Montréal; Centre Hospitalier de l'Université Laval, Ste-Foy; Centre Hospitalier Universitaire de Sherbrooke, Quebec; IWK Health Center, Halifax, Nova Scotia; Hospital for Sick Children, Toronto, Ontario, Canada; St Jude Children's Research Hospital, Memphis, TN; and AFLAC Cancer Center, Emory University and Children's Healthcare of Atlanta, Atlanta, GA Address reprint requests to Stéphane Barrette, 3175 Côte Ste-Catherine, bureau 3415, Montréal, Quebec, Canada, H3T 1C5; e-mail: stephane.barrette{at}umontreal.ca
Purpose: A significant increase in the incidence of neuroblastoma occurred among a 5-year birth cohort (May 1989 to April 1994) during an active urinary screening program for its early detection. We examined the postscreening incidence of neuroblastoma in the subsequent 5-year birth cohort (May 1994 to April 1999), with follow-up to 2002, to determine whether the incidence remained increased. Patients and Methods: We reviewed institutional records of patients diagnosed with neuroblastoma during the period from 1994 to 2002 who were born in 1994 to 1999 in the province of Quebec, as well as in the state of Minnesota and the province of Ontario, regions that had served as controls during the screening interval. We calculated and compared incidence rates during the 1994 to 2002 time period. Results: For the 5-year birth cohort as a whole, the rate of newly diagnosed neuroblastoma was higher in Quebec than in the control populations of Minnesota and Ontario (standardized incidence ratio, 1.34; 95% CI, 1.03 to 1.70). However, in years 4 and 5 of the interval, population-based incidence declined to the same levels as those seen in the control areas. Conclusion: The institution of a urinary screening program for neuroblastoma led to increased awareness of the diagnosis and an elevated rate of diagnosis even after the completion of the screening evaluation. However, this halo effect was transient, with diagnostic rates subsequently decreasing within the range seen in control populations.
Neuroblastoma is the most common extracranial solid tumor in young children, affecting one in 7,000 children younger than 5 years of age.1 The survival of patients depends on their age and the stage at diagnosis, with infants and early stages having the best prognosis.2,3 Because more than 85% of neuroblastomas secrete either homovanillic acid or vanillylmandelic acid, and because these specific catecholamine markers are easily measurable in urine,4 many screening programs have been undertaken during the last several decades with the purpose of decreasing mortality by detecting neuroblastoma in preclinical stages.5-7 The Quebec Neuroblastoma Screening Program (QNSP) was a US National Cancer Institute–funded research project designed to study screening for neuroblastoma. Specific aspects of the study have been published previously.8 Briefly, the study results showed an increased incidence of neuroblastoma in children age 0 to 11 months, with no decrease in incidence in older children, and no decrease in population-based mortality9 when compared with control populations within North America that had not been offered screening. During the screening time period, Quebec physicians also clinically diagnosed an increased number of early-stage patients. We hypothesized that this might have been due to the publicity surrounding the population eligible for screening and increased awareness on the part of examining physicians (the so-called halo effect). Before and during the period of screening (1989 to 1994), methods to improve compliance included publicity targeted at the population (especially new parents), as well as interventions aimed at medical professionals.10 Briefly, for medical professionals, these included letters to pediatricians, general practitioners, hospital nurseries, midwives, local community health centers, and regional health boards. Every nursery head nurse was contacted by telephone at the beginning of the screening period. In addition, articles were published in medical and nursing journals, seminars were given at professional meetings and in medical schools, and a videocassette describing the screening program was made available to nurseries and clinics. Furthermore, in March 1992, reminder letters were sent to physicians and clinics. To examine the durability of the halo effect, we undertook this evaluation of the impact of the screening program on the incidence of neuroblastoma in the postscreening interval.
The postscreening study population consisted of all births occurring in the provinces of Quebec and Ontario and the state of Minnesota for the period March 1, 1994 through April 30, 1999. The number of registered births, by month, was obtained from the governmental offices for the respective regions. Ascertainment of neuroblastoma diagnoses occurring within the 5-year postscreening birth cohort up to May 31, 2002, was conducted in a manner identical to that used for ascertainment of neuroblastoma diagnoses occurring during the 1989 to 1994 screened birth cohort.8 Within Quebec, this process consisted of systematic searches of medical records in the four provincial pediatric cancer centers and hospital tumor registries. Neuroblastoma diagnoses occurring in the state of Minnesota were identified through the population-based and pathology-based Minnesota Cancer Surveillance System, augmented and cross referenced with registrations through centers participating in the Children's Oncology Group. For the province of Ontario, information on neuroblastoma diagnoses was obtained from the Pediatric Oncology Group of Ontario. During the screening evaluation period (1989 to 1994), investigators used other available mechanisms such as provincial and state tumor registries, Children's Cancer Group and Pediatric Oncology Group registries, and for Ontario and Quebec, Statistics Canada data to search for patients. All but a few patients had already been known to investigators using institutional registries. Hence, only institutional registries were used for the period after the screening interval reported here (1994 to 1999). Follow-up was complete through April 30, 2002, providing a minimum follow-up of 3 years and a maximum follow-up of 8 years. This is slightly longer than the 15- to 75-month follow-up period analyzed in the comparison of the incidence of neuroblastoma in the screened Quebec population and the contemporary unscreened control population.8 To evaluate the occurrence of neuroblastoma within the postscreening study populations, the person-years of observation were compiled within 12-month age strata. Using 1-year age-specific incidence rates from the US Surveillance, Epidemiology, and End Results (SEER) program,1 we calculated standardized incidence ratios (SIRs) and corresponding 95% CIs. In addition, comparisons were carried out using the incidence rates from the province of Ontario and the state of Minnesota for their respective 1994 to 1999 birth cohorts, which were then applied to calculate expected cases within the province of Quebec.
A total of 412,919 births occurred in the province of Quebec in the 5-year period after cessation of the neuroblastoma screening program. Through the period of follow-up (May 31, 2002), there were 1,877,883 person-years of follow-up. Sixty-two diagnoses of neuroblastoma occurred among the postscreening birth cohort, including 28 diagnoses in the first 12 months of life, 14 diagnoses between 12 and 23 months, 11 diagnoses for 24 to 35 months, and nine diagnoses for  36 months of age. Births within the control populations of Ontario and Minnesota were 693,611 and 321,229, respectively, reflecting total person-years of follow-up of 3,136,539 and 1,436,446, respectively. One hundred thirteen incidence cases of neuroblastoma were diagnosed in Ontario (n = 78) and Minnesota (n = 35). Table 1 lists the age-specific standardized incidence ratios for patients diagnosed in the province of Quebec in the 5-year postscreening interval. Expected occurrences were calculated using age-specific rates from national US incidence (SEER registry1), as well as the calculated incidence within the control populations of Ontario and Minnesota. When compared with US SEER data, the SIR for Quebec was 1.19 (95% CI, 0.91 to 1.51). However, the control populations used during the screening interval (from Ontario and Minnesota) had lower rates of neuroblastoma in the 1994 to 1999 time period (SIR, 0.87; 95% CI, 0.61 to 1.18 v SIR, 0.90; 95% CI, 0.71 to 1.11, respectively). Comparing the Quebec rate with that of the combined Minnesota and Ontario populations yielded a significantly higher SIR (1.33; 95% CI, 1.02 to 1.68).
The SIRs according to birth cohort are shown in Figure 1, which demonstrates the prescreening (1977 to 1986) incidence rate compared with the Greater Delaware Valley; the increased incidence ratio compared with Minnesota and Ontario during the screening interval (May 1989 to April 1994); and the increased but not statistically significantly elevated incidence ratios for neuroblastoma in Quebec, again compared with the combined Minnesota and Ontario population for births in the first 3 years after the screening interval. However, the incidence ratio decreased in years 4 and 5, suggesting an attenuation of the halo effect.
The QNSP was designed to test the impact of the introduction of a urinary screening program for the catecholamine metabolites homovanillic acid and vanillylmandelic acid on the incidence and mortality of neuroblastoma in the pediatric population. In the 10-year prescreening period (1977 to 1986), for the population younger than 5 years of age, the incidence rate ratio, comparing Quebec with a large control group (the Greater Delaware Valley) was 0.98.11 During the screening interval, May 1989 to April 1994, the incidence rate ratio increased to 2.11 and was even higher in children younger than 12 months of age when compared with the control populations of Ontario and Minnesota, which historically had incidence rates of neuroblastoma similar to those in the Greater Delaware Valley. However, population-based mortality remained unchanged.9 We observed the Quebec and control populations born in the 5-year interval after the screening period. Compared with the control populations, the incidence in Quebec remained elevated among the first three annual birth cohorts after termination of the neuroblastoma screening program (rate ratios of 1.68, 2.16, and 2.06, respectively), which is in contrast to the subsequent two birth cohorts (1997 to 98 and 1998 to 1999), in which the incidence was closer to or below that observed in Ontario and Minnesota (rate ratios of 0.63 and 1.21, respectively). This suggests an attenuation of any halo effect that might have been induced by the publicity surrounding urine screening for neuroblastoma, and no change in true population-based incidence. Similarly, it would be of interest for Japanese investigators to study national incidence rates after their long experience with mandatory 6-month urinary screening for neuroblastoma, now that screening has been halted. More broadly, publicity surrounding a screening intervention may have the desired effect of increasing the participation in that intervention. In addition, practitioners may be sensitized to early signs of disease and may increase the clinical diagnostic rate. Developing technologies may also increase the diagnosis of asymptomatic disease, as is seen in antenatally diagnosed neuroblastoma, or incidental neuroblastomas diagnosed during investigation of urinary tract infections, for example. If the screening procedure truly leads to early intervention and improved outcome, this is a desirable goal. However, this must be assessed carefully by the evaluation of population-based incidence and mortality before concluding that screening has been a success, based solely on patient fatality rates. Problems of overdiagnosis and possible overtreatment causing unnecessary complications in patients with disease that would not become symptomatic, as seen in the QNSP,12 also arise in much more common malignancies in adults, such as prostate cancer13 and, to a lesser degree, breast cancer.14 The concern of balancing the benefit of early detection compared with the risk of overdiagnosis and possible overtreatment also led to the initiation of the large prostate, lung, colorectal, and ovarian randomized screening study, for which follow-up and analysis are ongoing.15 In summary, the demonstrated halo effect in the QNSP during the screening period, leading to increased clinical diagnosis presumably by increased community and professional awareness, extinguished during a 5-year period. Hence, the overall incidence of neuroblastoma in Quebec is once again similar to that observed in other parts of North America.
The author(s) indicated no potential conflicts of interest.
Conception and design: Mark L. Bernstein, Leslie L. Robison, William G. Woods Provision of study materials or patients: Stéphane Barrette, Mark L. Bernstein, Yvan Samson, Josée Brossard, Sheila Weitzman Collection and assembly of data: Stéphane Barrette, Mark L. Bernstein, Leslie L. Robison, Yvan Samson, Josée Brossard, Sheila Weitzman, William G. Woods Data analysis and interpretation: Stéphane Barrette, Mark L. Bernstein, Leslie L. Robison, William G. Woods Manuscript writing: Stéphane Barrette, Mark L. Bernstein, Leslie L. Robison, William G. Woods Final approval of manuscript: Stéphane Barrette, Mark L. Bernstein, Leslie L. Robison, Yvan Samson, Josée Brossard, Sheila Weitzman, William G. Woods
We thank the clinical research associates of the participating institutions (Doreen Lalonde, Louise Renaud, Patricia Campion, and Aline Blais) and Timothy Byrne at the Project Coordinating Center.
Supported by Grant No. CA46907) from the National Institutes of Health. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Ries LAG, Eisner MP, Kosary CL, et al: SEER Cancer Statistics Review, 1975-2002. Bethesda, MD, National Cancer Institute, 2002 2. Shimada H, Umehara S, Monobe Y, et al: International neuroblastoma pathology classification for prognostic evaluation of patients with peripheral neuroblastic tumors: A report from the Children's Cancer Group. Cancer 92 : 2451 -2461, 2001[CrossRef][Medline] 3. Ikeda H, Iehara T, Tsuchida Y, et al: Experience with International Neuroblastoma Staging System and Pathology Classification. Br J Cancer 86 : 1110 -1116, 2002[CrossRef][Medline] 4. Tuchman M, Lemieux B, Auray-Blais C, et al: Screening for neuroblastoma at 3 weeks of age: Methods and preliminary results from the Quebec Neuroblastoma Screening Project. Pediatrics 86
: 765
-773, 1990 5. Berthold F, Baillot A, Hero B, et al: Which cases are found and missed by neuroblastoma screening at 1 year? Results from the 1992 to 1995 study in three Federal States of Germany. J Clin Oncol 17
: 1200
-1207, 1999 6. Honjo S, Doran HE, Stiller CA, et al: Neuroblastoma trends in Osaka, Japan, and Great Britain 1970-1994, in relation to screening. Int J Cancer 103 : 538 -543, 2003[CrossRef][Medline] 7. Carlsen NL: Neuroblastoma: Epidemiology and pattern of regression: Problems in interpreting results of mass screening. Am J Pediatr Hematol Oncol 14 : 103 -110, 1992[Medline] 8. Woods WG, Tuchman M, Robison LL, et al: A population-based study of the usefulness of screening for neuroblastoma. Lancet 348 : 1682 -1687, 1996[CrossRef][Medline] 9. Woods WG, Gao RN, Shuster JJ, et al: Screening of infants and mortality due to neuroblastoma. N Engl J Med 346
: 1041
-1046, 2002 10. Campion P, Woods WG, Lemieux B: Compliance in a screening program for neuroblastoma. Prev Med 27 : 590 -596, 1998[CrossRef][Medline] 11. Bernstein ML, Leclerc JM, Bunin G, et al: A population-based study of neuroblastoma incidence, survival and mortality in North America. J Clin Oncol 10 : 323 -329, 1992[Medline] 12. Barrette S, Bernstein ML, Leclerc JM, et al: Treatment complications in children diagnosed with neuroblastoma during a screening program. J Clin Oncol 24
: 1542
-1545, 2006 13. Etzioni R, Penson DF, Legler JM, et al: Overdiagnosis due to prostate-specific antigen screening: Lessons from U.S. prostate cancer incidence trends. J Natl Cancer Inst 94
: 981
-990, 2002 14. Gotzsche PC, Nielsen M: Screening for breast cancer with mammography. Oxford, United Kingdom, Cochrane Library, CD001877, 4, 2006 15. Prorok PC, Andriole GL, Bresalier RS et al: Design of the prostate, lung, colorectal and ovarian (PLCO) cancer screening trial. Control Clin Trials 21 : 273S -309S, 2000[CrossRef][Medline] Submitted April 16, 2007; accepted August 10, 2007.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
