|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Decreasing Late Mortality Among Five-Year Survivors of Cancer in Childhood and Adolescence: A Population-Based Study in the Nordic CountriesByFrom the Departments of Cancer Epidemiology and Pediatrics, University Hospital, Lund; and Swedish Cancer Registry, Stockholm, Sweden; Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark; Cancer Registry of Norway, Oslo, Norway; Icelandic Cancer Registry, Reykjavik, Iceland; and Finnish Cancer Registry, Helsinki, Finland. Address reprint requests to Torgil R. Möller, MD, PhD, Department of Cancer Epidemiology, Southern Swedish Tumor Registry, University Hospital, SE-221 85 Lund, Sweden; email: torgil.moller{at}cancerepid.lu.se
PURPOSE: To assess the risk of death in patients who survive more than 5 years after diagnosis of childhood cancer and to evaluate causes of death in fatal cases. PATIENTS AND METHODS: This was a population-based study in the five Nordic countries (Denmark, Finland, Iceland, Norway, and Sweden) using data of the nationwide cancer registries and the cause-of-death registries. The study cohort included 13,711 patients who were diagnosed with cancer before the age of 20 years between 1960 and 1989 and who survived at least 5 years from diagnosis. By December 31, 1995, 1,422 patients had died, and death certificates were assessed in 1,402. Standardized mortality ratios (SMRs) for validated causes of death were calculated based on 156,046 patient-years at risk. RESULTS: The overall SMR was 10.8 (95% confidence interval [CI], 10.3 to 11.5), mainly due to high excess mortality from the primary cancer. SMR for second cancer was 4.9 (95% CI, 3.9 to 5.9) and was 3.1 (95% CI, 2.8 to 3.5) for noncancer death. The pattern of causes of death varied markedly between different groups of primary cancer diagnoses and was highly dependent on time passed since diagnosis. Overall late mortality was significantly lower in patients treated during the most recent period of time, 1980 to 1989, compared with those treated from 1960 to 1979 (hazard ratio, 0.61; 95% CI, 0.54 to 0.70), and there was no increase in rates of death due to cancer treatment. CONCLUSION: Long-term survivors of childhood cancer had an increased mortality rate, mainly dying from primary cancers. However, modern treatments have reduced late cancer mortality without increasing the rate of therapy-related deaths.
TREATMENT OF childhood cancer has become more effective, resulting in more patients surviving at least 5 years from diagnosis. Practically all children in Western countries are initially treated for their cancer in departments of pediatrics or centers for pediatric oncology, but data on long-term follow-up after 5 years from diagnosis are sparse. A recent survey among pediatric oncology institutions in the United States1 revealed that less than half of these institutions had a mechanism for following up adult survivors, and only 15% had established a formal database for such individuals. In the Nordic countries, there is no general policy for the long-term follow-up of adult survivors of childhood cancer. The overall pattern of late mortality among childhood cancer survivors has been studied in only a few settings. In two population-based studies from the United Kingdom, the Childhood Cancer Research Group analyzed the causes of death in 5-year survivors of childhood cancer treated before 19712 and in those treated between 1971 and 1985.3 There was a reduction in the risk of dying from recurrent tumor and a slight increase in the risk of dying from treatment-related effects among those treated between 1971 and 1985, compared with those treated before 1971. In the 1971 to 1985 study, the risk of death from non–neoplastic disorders was four-fold greater than in the general population.3 In the United States, a single-institution investigation at St Jude Children’s Research Hospital of patients who had survived at least 5 years from diagnosis showed a significant reduction of deaths from recurrent primary cancer in patients diagnosed from 1971 to 1983 as compared with those diagnosed from 1962 to 1970, an insignificant increase of second malignancies, and no differences in treatment complications, unintentional injuries, or suicide.4 However, the study cohort included a low proportion of children with brain tumors and a high proportion of patients with leukemia because of the selection of study subjects. Recently, Green et al5 showed that 15-year survivors of childhood cancer had excess mortality due to second malignant neoplasms and cardiac disease, with no decrease in the late mortality rate in the more recent treatment era (1971 to 1984). Some investigators6 express apprehension that modern intensive therapy may lead to increased mortality later in life from causes other than the primary cancer. In official mortality statistics, the underlying cause of death should be defined as "the disease (or injury), which initiated the train of morbid events leading directly to death."7 In cases with a malignancy, this usually means that the cancer is selected as the underlying cause of death, leading to a possible overestimation of tumor deaths and underestimation of deaths due to other causes, eg, side effects of treatment. From the clinical point of view, information on such serious side effects is of utmost importance and might necessitate changes in primary cancer treatment. Such information is, however, because of the reasons given, not so obvious in official mortality statistics. It was therefore decided to undertake a study of the most probable causes of death in long-term survivors of childhood cancer, based on a review of the corresponding death certificates. The aim of the present study was thus to assess the long-term mortality in a population-based cohort of 13,711 patients diagnosed with childhood cancer in the Nordic countries from 1960 to 1989 and surviving at least 5 years after diagnosis.
Patients Patient data were obtained from the nationwide and population-based cancer registries and cause-of-death registries in the five Nordic countries (Denmark, Finland, Iceland, Norway, and Sweden). The basic population for the study was formed of 27,270 individuals (15,074 males and 12,196 females) diagnosed with a malignant tumor before 20 years of age during 1960 to 1989. The study cohort consisted of all patients alive 5 years after the first diagnosis of cancer and comprised 13,711 patients (7,153 males and 6,558 females). The patients were followed up through December 31, 1995, with respect to vital status, emigration, and cause of death. All primary malignant tumors were classified according to the International Classification of Diseases for Oncology8 and grouped according to Birch and Marsden’s classification scheme for childhood cancer, proposed by the International Agency for Research on Cancer (IARC group).9 Table 1 shows the distribution of the patients according to the decade of diagnosis of the first cancer. As can be seen, the number of 5-year survivors increases with time, and the proportion of different IARC groups changes. During the same three decades, the number of newly diagnosed cases as well as the proportion of different IARC groups at diagnosis were rather stable (data not shown). There were 4,012 individuals diagnosed at the age of 0 to 4 years, 2,426 individuals diagnosed at the age of 5 to 9 years, 2,669 diagnosed at the age of 10 to 14 years, and 4,604 diagnosed at the age of 15 to 19 years.
At the last follow-up (December 31, 1995), 1,422 (832 males and 590 females) patients (10.4%) were dead, while 165 individuals (1.2%) were lost to follow-up due to emigration. Long-term survivors of Hodgkin’s disease experienced the highest proportion of fatalities (19.4%), followed by patients with leukemia (13.6%) and CNS tumors (13.6%). The lowest proportion of deaths occurred in patients with renal tumors (4.2%), retinoblastoma (4.5%), and germ cell tumors (4.6%). The median follow-up after diagnosis for the entire cohort was 14.9 years (range, 5.0 to 35.9 years). The shortest follow up was for hepatic tumors (11.2 years) and leukemia (11.9 years), and the longest, for retinoblastoma (20.1 years).
Assessment of the Causes of Death Further, all validated death causes were classified in one of the following five categories: Ca1, death caused by the first cancer (progression, recurrence, or metastasis); Ca2, death caused by a second or subsequent primary cancer; Th1, death related to the therapy of the first tumor without signs of tumor growth; Th2, death related to the therapy of the second or subsequent tumor without signs of tumor growth; and Un, death due to causes seemingly unrelated to cancer or its therapy.
Statistical Methods Sex, calendar year, and age-specific (1- or 5-year classes, as available) mortality rates for causes of death according to the Nordic abbreviated list were available on a population basis from all five Nordic countries from 1971 on. Missing rates for 1994 (two countries) and 1995 (three countries) were replaced by the rates of the previous year. Standardized mortality ratios (SMRs) were calculated for each of the 15 major diagnostic groups based on the abbreviated Nordic list. Calculations were based on validated causes of death. In addition, SMRs for all causes of death, death caused by a second or subsequent primary cancer (category Ca2), and noncancer death (categories Th1 + Th2 + Un) were calculated. Since national mortality figures stratified according to the Nordic abbreviated list were not uniformly available for 1965 to 1970, the patients who died (n = 99) or emigrated (n = 8) before January 1971 were excluded from the SMRs analysis. Sixteen patients with missing death certificates were censored at time of death with respect to all SMR calculations, except when all causes of death were considered. All confidence intervals (CIs) had a coverage probability of 95%.
Overall Mortality Estimated overall mortality for the 5-year survivors was 0.10 (95% CI, 0.09 to 0.10) at 15 years after diagnosis, and 0.14 (95% CI, 0.13 to 0.14) at 25 years after diagnosis. Corresponding expected mortality rates were 0.01 and 0.02, respectively, showing that most of the deaths in the study cohort represented excess mortality relative to the general population. Higher mortality rates were observed among males (log-rank P < .0001). At 25 years, mortality was 0.12 in females, compared with 0.15 in males. For patients diagnosed with cancer during the first 5 years of life, mortality was lower than in the older age groups (log-rank P = .0001), estimated as 0.11, compared with 0.15 at 25 years after diagnosis. The cumulative mortality decreased significantly in later decades of primary cancer diagnosis (Fig 1, log-rank P < .0001). This was especially marked for patients diagnosed during the most recent period of time, ie, 1980 to 1989. The pattern was similar for both sexes (not shown). In a Cox proportional hazards regression, stratified by age at diagnosis (four groups) and adjusted for sex, the overall death hazard ratio (HR) comparing patients diagnosed from 1980 to 1989 with those diagnosed from 1960 to 1979 was 0.61 (95% CI, 0.54 to 0.70). In separate analyses, the decreased mortality was most pronounced in patients with leukemia (HR = 0.34; 95% CI, 0.27 to 0.43) and with Hodgkin’s disease (HR = 0.35; 95% CI, 0.23 to 0.52), but for CNS tumors the changes were not so apparent (HR = 0.81; 95% CI, 0.65 to 1.01).
Validated Causes of Death The causes of death could be validated in 1,402 out of 1,422 patients. As listed in Table 2, 976 individuals (69.6%) succumbed to the first cancer (category Ca1), and 99 (7.1%) died because of a second or subsequent primary cancer (category Ca2). In 155 patients (11.1%), death was due to causes related to therapy of the first cancer (category Th1); in five patients (0.4%), to therapy of the second or subsequent primary cancer (category Th2); and in 167 patients (11.9%), no obvious relation between death and neoplastic disease or its therapy was found (category Un). Out of 1,402 patients in whom the cause of death could be validated, 1,208 (86.2%) died because of tumor activity, according to the official statistics, whereas the validation concluded that 1,075 patients (76.7%) actually died from tumors. In 1,060 out of 1,075 validated tumor deaths, the official cause was also tumor (sensitivity, 98.6%; CI, 97.7 to 99.2), whereas only 179 out of 327 validated nontumor deaths were officially classified as nontumor (specificity, 54.7%; CI, 49.2 to 60.2). Discordance in classification thus occurred in 163 cases.
Validated Cause of Death in Relation to Primary Diagnosis The pattern of causes of death varied markedly among different groups of primary cancer diagnoses (Table 2). Second or subsequent malignant neoplasm (SMN) most often ended the lives of patients with retinoblastoma and renal tumors. Out of eight cases of SMN in retinoblastoma, three were bone tumors and three, malignant melanomas. No consistent pattern was seen in seven SMNs occurring in renal tumors. Out of 29 fatal SMNs in patients with Hodgkin’s disease, seven occurred in the gastrointestinal tract, seven were leukemia (mostly acute myeloid), and three were breast carcinomas. Among eight SMNs occurring in leukemia, five were brain tumors. Therapy-related death was most prevalent in patients with Hodgkin’s disease, renal tumors, and leukemia (Table 2). In Table 3, all noncancer causes of death (categories Th1 + Th2 + Un) are shown according to the primary cancer diagnosis (IARC group). For patients with CNS tumors, the dominating causes of death were diseases of the respiratory system, including pneumonia, diseases of the CNS (including cerebrovascular disease), and accidents, including motor vehicle accidents. In those experiencing Hodgkin’s disease, most died of diseases of the respiratory tract, heart, and circulatory system. Among patients with leukemia, death was most often caused by diseases of the respiratory tract. In total, accidents and diseases of the respiratory tract, heart, and CNS, including cerebrovascular disease, caused 69% of all noncancer deaths (Table 3).
SMR for Validated Causes of Death As can be seen in Table 4, the SMR based on validated causes of death for the period from 1971 to 1995 was 10.8 for all causes, with 4.9 for death due to second or subsequent cancer and 3.1 for all noncancer causes. There was a difference between sexes, SMRs being lower for males for all death causes and noncancer death. The overall SMR decreased with age at diagnosis and with increasing follow-up time (Table 4).
Table 5 shows the SMRs for validated cause of death for all causes, except first cancer, grouped in 15 major diagnostic groups based on the Nordic abbreviated list of causes of death. Markedly increased SMR values were observed for diseases of the respiratory system, including pneumonia, followed by diseases of the genitourinary system, infectious diseases, diseases of the heart and circulatory system, SMNs, and diseases of the CNS, including cerebrovascular disease. The SMR was not increased for alcoholism or suicide (Table 5).
Cause of Death According to Length of Follow-Up Fifteen years after diagnosis, the estimated cumulative mortality due to first cancer was 7.2%; due to second cancer, 0.5%; and due to all noncancer causes, 1.9%. Corresponding figures 25 years after diagnosis were 8.8%, 1.2%, and 3.6%, respectively (Fig 2).
The influence of the length of survival on the pattern of causes of death was also dependent on primary diagnoses. For all patients, the proportion of deaths from the first tumor decreased from 82% during 5 to 10 years after diagnosis to 34% 20 years or more after diagnosis (Fig 3a). On the contrary, the proportion of deaths from second malignant tumor increased at the same points of time from 3% to 22%. The proportion of therapy-related deaths was relatively constant, approximately 10%, while the proportion of unrelated death causes increased from 5% to 32% (Fig 3a). The observed pattern varied markedly between the different IARC groups, as shown in Figure 3b (Hodgkin’s disease), c (leukemia), and d (CNS tumors).
Cause-Specific Death Hazards by Time Period of Primary Diagnosis Table 6 shows that, assuming proportional death hazards, a decrease in the hazard of death as a result of first tumor was observed for patients diagnosed from 1980 to 1989, compared with the two preceding decades (estimated HR = 0.60). The decrease was most pronounced for patients with leukemia (HR = 0.34) and Hodgkin’s disease (HR = 0.28). However, the change was significant also for CNS tumors (HR = 0.74) and all other tumors (HR = 0.70). For death due to second malignancy (category Ca2), the hazards were not significantly different between the two periods. For therapy-related mortality (categories Th1 + Th2), the death hazard was significantly lower for patients with leukemia diagnosed from 1980 to 1989 (HR = 0.34) and almost significantly lower for all patients (HR = 0.67; CI, 0.45 to 1.00). For all noncancer deaths (categories Th1 + Th2 + Un), the hazard was significantly lower for those patients diagnosed from 1980 to 1989 (HR = 0.70) and especially for patients with leukemia (HR = 0.39) and Hodgkin’s disease (HR = 0.39) (Table 6).
The material in this study is derived from the basic population of 27,270 childhood cancer patients diagnosed in the Nordic countries during three decades from 1960 to 1989, with few patients lost to follow-up. Thirteen thousand, seven hundred eleven patients (50.3%) are known to have survived at least 5 years after diagnosis. However, survival was not equal to cure, because approximately 10% of the 5-year survivors died during the subsequent follow-up. In relation to mortality in the general population, practically all these deaths constitute excess mortality. We have found a clear decrease in late total mortality over time. The death hazard in 5-year survivors diagnosed in the eighties was estimated to be 40% less than that of those diagnosed in the two previous decades. These findings are in line with the reports of Robertson et al3 and Hudson et al.4 Due to the rules of the official mortality statistics as applied in the Nordic countries,7,10 death attributable to the first malignant diagnosis can be overestimated, and the direct or intervening antecedent cause of death, which might reflect therapy-related morbidity and mortality, can be underestimated. In addition, in this population-based study, covering a large area and prolonged period of time, the excellent follow-up of all patients was counterbalanced by the lack of pertinent clinical data regarding treatments of the cancers. We tried to circumvent these obstacles by validating the cause of death using death certificates together with the data from cancer registries and cause-of-death statistics in all the Nordic countries. A comparison of the official causes of death with the validated causes of death, as carried out in this study, showed that 12% (148 of 1208) of fatalities attributed to the first malignancy were in fact noncancer deaths. Fifteen cases officially recorded as death due to malformation were at validation reclassified as deaths caused by primary tumors; almost all cases were neurofibromatosis type I, which was explained by the coding praxis. SMR for all death causes was 10.8, the main reason for the excess mortality being recurrence or progression of the primary tumor (total, 70%). Even more than 20 years after diagnosis, one third of fatalities were due to the primary tumor. At that time of follow-up, approximately the same proportion of deaths was due to causes seemingly unrelated to cancer, and about one fifth to a second (or subsequent) malignant tumor. For females, overall mortality was distinctly lower than for males, but the SMR was almost 60% greater (14.6 v 9.2). This is explained by the much lower expected mortality among females. SMR for death due to SMN was 4.9, with a pattern of SMN causing fatalities differing from the usual incidence of SMN after childhood cancer. This was perhaps most apparent in patients after primary diagnoses of Hodgkin’s disease. Although in the present study, three (10%) out of 29 fatal SMNs were breast carcinomas, this diagnosis accounted for 26% (16 out of 62) of the total amount of SMN occurring in the Nordic cohort.15 On the contrary, SMNs in the gastrointestinal tract were causing relatively higher mortality, 24% (seven of 29 patients) compared with an 11% (seven of 62) occurrence in the same cohort.15 An interesting pattern of fatal SMN cases was also observed in patients with retinoblastoma, who died equally frequently of malignant melanomas and bone tumors. While in patients with Hodgkin’s disease the pattern of SMNs is, as far as is known at present, largely dependent on the treatment modality, in retinoblastoma, on the contrary, the development of cutaneous melanomas is claimed to be independent of the radiation therapy.16 It is noteworthy that the death HR from SMN apparently did not increase during the most recent decade studied, suggesting that the treatment used in the eighties was not more carcinogenic than previously used treatments. SMR for validated noncancer death was 3.1, which was similar to the four-fold excess found by Robertson et al.3 The highest figures were observed for the diseases of the respiratory tract, including pneumonia (34.9), diseases of the genitourinary tract, (11.1, based on a few cases), infectious diseases (8.6), and diseases of the heart and of the circulatory system (5.8). Suicides and deaths from alcoholism were not more frequent than expected. Noncancer deaths were clearly underestimated in the official statistics, especially diseases of the respiratory tract, including pneumonia, diseases of the heart and of the circulatory system, diseases of the CNS including cerebrovascular disease, and infectious diseases. Although no details on treatment were available in the individual cases, deaths that could be attributed to therapy were classified based on knowledge of the treatment practice prevailing at the time of diagnosis. For example, combination chemotherapy for leukemia was introduced in the early seventies and intensified during the early eighties, and the use of radiotherapy to the CNS, introduced in the seventies, diminished in the late eighties.17 For Hodgkin’s disease, involved-field radiotherapy was replaced by the extended field techniques (mantel and inverted-Y) in the seventies18,19 simultaneous with the introduction of combination chemotherapy (mechlorethamine, vincristine, procarbazine, prednisone). On the other hand, CNS tumors did not, as a rule, receive any chemotherapy, and postoperative radiotherapy was given only to malignant tumors. In the present study, it was sometimes difficult to decide if the death was due to the treatment effect or was unrelated to the cancer and its treatment. Due to the study design, we were not able to analyze in detail the effect of different treatment modalities or disease recurrence on late mortality, which was reported recently by Green et al.5 Nevertheless, the pattern of HRs was similar whether therapy-related deaths only or therapy-related deaths combined with deaths seemingly unrelated to the cancer diagnosis were taken into account and showed that late noncancer mortality was lower in the Nordic countries in the eighties, compared with the two previous decades. This result differs from the study of Robertson et al,3 who found a slight increase from 1% to 2% in the risk of dying from treatment-related effects between patients diagnosed from 1940 to 1970 and those diagnosed from 1971 to 1985, and from Hudson et al,4 who did not identify any significant change in late death secondary to treatment complications in patients diagnosed from 1971 to 1985, as compared with those diagnosed from 1962 to 1970. However, it is to be remembered that follow-up for the patients diagnosed in the later part of our study is relatively short, and that mortality can increase for these patients both from noncancer causes and from SMNs. As in the above quoted studies,3,4 we found a significant reduction of mortality from the first malignant neoplasm in the later decades of the study, and this decrease must be considered definite. The decrease was most pronounced for patients with Hodgkin’s disease and leukemia, while for CNS tumors, only a slight improvement was observed. Theoretically, late mortality could be further reduced by more effective treatment of the primary cancer and by preventing therapy-related fatalities. If the etiologic prevention is not possible without jeopardizing the treatment results, then the solution must be the use of effective surveillance programs for long-term childhood cancer survivors, especially for those with known risk factors for late morbidity and mortality. In accordance with previous report,20 one such high-risk group in our study was patients with Hodgkin’s disease, who experienced the highest proportion of treatment-related fatalities, including secondary malignancies, diseases of the respiratory tract, and diseases of the heart and circulatory system. Another group emerging in our study was 5-year survivors of CNS tumors, in whom by far the most common cause of noncancer deaths (30 of 113; 26.5%) was diseases of the respiratory system, including pneumonia. Taking into account the excellent results of the modern treatment of most childhood malignancies, decreasing late morbidity and mortality of the cured patients becomes a growing challenge for the health system.
Supported by research grants from the Swedish Childhood Cancer Foundation, the University Hospital Funds in Lund, and the Malmöhus County Council, Sweden. Harald Anderson, PhD, provided valuable statistical expertise during the early part of the study. The skillful assistance of Gertrud Andersson is also highly appreciated.
1. Oeffinger KC, Eshelman DA, Tomlinson GE, et al: Programs for adult survivors of childhood cancer. J Clin Oncol 16: 2864-2867, 1998[Abstract] 2. Hawkins MM, Kingston JE, Kinnier Wilson LM: Late deaths after treatment for childhood cancer. Arch Dis Child 65: 1356-1363, 1990[Abstract]
3.
Robertson CM, Hawkins MM, Kingston JE: Late deaths and survival after childhood cancer: Implications for cure. BMJ 309: 162-166, 1994
4.
Hudson MM, Jones D, Boyett J, et al: Late mortality of long-term survivors of childhood cancer. J Clin Oncol 15: 2205-2213, 1997
5.
Green DM, Hyland A, Chung CS, et al: Cancer and cardiac mortality among 15-year survivors of cancer diagnosed during childhood or adolescence. J Clin Oncol 17: 3207-3215, 1999 6. Nicholson HS, Fears TR, Byrne J: Death during adulthood in survivors of childhood and adolescent cancer. Cancer 73: 3094-3102, 1994[Medline] 7. World Health Organization : Manual of the International Statistical Classification of Diseases—Injuries and Causes of Death, 1975 Revision. Geneva, Switzerland, World Health Organization, 1977 8. World Health Organization : ICD-O International Classification of Diseases for Oncology ( ed 1 ). Geneva, Switzerland, World Health Organization, 1976 9. Birch JM, Marsden HB: A classification scheme for childhood cancer. Int J Cancer 40: 620-624, 1987[Medline] 10. Johansson LA, Bille H, Ahonen H, et al: Cause-of-Death Statistics, in Nordic Medico Statistical Committee (ed): Health Statistics in the Nordic Countries 1997. Copenhagen, Denmark, NOMESCO, 1999, pp 175-208 11. The National Board of Health and Welfare : Causes of Death, 1994. Stockholm, Socialstyrelsen, Epidemiologiskt Centrum, 1996 12. Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY, John Wiley & Sons, 1980 13. Hakulinen T: Cancer survival corrected for heterogeneity in patient withdrawal. Biometrics 38: 933-942, 1982[Medline] 14. Hosmer DW, Lemeshow S: Applied Logistic Regression. New York, NY, John Wiley & Sons, 1989
15.
Sankila R, Garwicz S, Olsen JH, et al: Risk of subsequent malignant neoplasms among 1,641 Hodgkin’s disease patients diagnosed in childhood and adolescence: A population-based cohort study in the five Nordic countries. J Clin Oncol 14: 1442-1446, 1996 16. Abramson D: Second nonocular cancers in retinoblastoma: A unified hypothesis—The Franceschetti Lecture. Ophthalmic Genet 20: 193-204, 1999[Medline] 17. Gustafsson G, Kreuger A, Clausen N, et al: Intensified treatment of acute childhood lymphoblastic leukaemia has improved prognosis, especially in non–high-risk patients: The Nordic experience of 2,648 patients diagnosed between 1981 and 1996. Acta Paediatr 87: 1151-1161, 1998[Medline] 18. Svahn-Tapper G, Baldetorp L, Landberg T: Mantle treatment of Hodgkin’s disease: Results and side effects. Acta Radiol Scand 15: 369-386, 1976 19. Mercke C, Landberg T, Svahn-Tapper G: Hodgkin’s disease irradiated with the inverted-Y technique. Acta Radiol Scand 20: 81-89, 1981 20. Hudson MM, Poquette CA, Lee J, et al: Increased mortality after successful treatment for Hodgkin’s disease. J Clin Oncol 16: 3592-3600, 1998[Abstract] Submitted November 28, 2000; accepted April 3, 2001. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
