Causes of Mortality in Survivors of Childhood Cancer

1
Causes of Mortality in Survivors of Childhood
Cancer

• Ann Mertens, Ph.D.
• Department of Pediatrics
• University of Minnesota

2
Background

• Estimated that over 8000 children diagnosed with
  cancer in the U.S. each year

• Current survival rate over 70

3
Trends in Childhood Cancer 5-year survival (age
under 15) Cancer
4
Selected Late Effects of Childhood Cancer

• Growth
• Physical impairment
• Gonadal Function
• Fertility/Offspring
• Subsequent Malignancy
• Cardiotoxicity

• Pulmonary Function
• Neuropsychologic
• Psychosocial
• Liver Function
• Thyroid
• Gastrointestinal

5
Early Suggestion of Late Mortality
The Course of Five-Year Survivors of Cancer in
Childhood Li et al. J Pediatrics,
1978.

• Data from 4 SEER sites, DFCI
• Diagnosis between 1947-1960
• 2232 5-year Survivors / 216 Deaths (9.7)
• Steepest decline in survival 5-9 years post-dx

6
Previous Studies on Late Mortality
Late Deaths and Survival after Childhood Cancer
Implications for Cure. Robertson et al. BMJ,
1994.

• Population-based study from Britain
• Diagnosis between 1971-1985
• 9080 5-year survivors / 793 Deaths (8.7)
• 74 of deaths attributed to recurrence
• 22 of deaths attributed to treatment

7
Previous Studies on Late Mortality
Late Mortality of Long-Term Survivors of
Childhood Cancer. Hudson et al., J.
Clinical Onc., 1997.

• St. Jude Childrens Research Hospital
• Diagnosis between 1960-1970, 1971-1983
• 2053 5-year survivors / 238 deaths (11.6)
• Death from recurrence decreased in recent era
• Death from 2nd cancer increased in recent era

8
Two Contemporary Studies on Late Mortality in
Childhood Cancer Survivors
Mertens et al Moller et al Country USA Nordi
c Type of study Hospital Population Years of
diagnosis 1970-1986 1960-1989 5-year
survivors 20,227 13,711 Deaths 2030 (10)
1422 (10) SMR 10.8 10.8 Deaths
recurrence 67 70 - treatment-related
23 18 - non-cancer related 10 12 J
Clin Oncol, 2001
9
Very Late Mortality Experience in Five-Year
Survivors ofChildhood and Adolescent CancerAC
Mertens, JP Neglia, Q Liu, K Wasilewski, LL
Robison, Y Yasui,
10
Childhood Cancer Survivor Study (CCSS)

• Retrospective cohort study
• 20,000 childhood cancer survivors
• 25 institutions in the U.S. and Canada
• Hypotheses related to sequelae of cancer
  treatment

11
CCSS Study Overview

• Eligibility criteria
• 5-Year Survival
• Leukemia, Lymphoma, CNS, Bone, Wilms, NBL,
  Soft-tissue sarcoma
• Diagnosis 1970-1986

12
CCSS Eligible Population
Diagnosis Number () Leukemia 6665 (33)
CNS Tumors 2836 (14) Hodgkin Disease 2718
(13) Soft Tissue Sarcoma 1815 (9)
Kidney (Wilms) 1684 (8) Bone
Tumors 1731 (8) Non-Hodgkin Lymphoma 1488
(7) Neuroblastoma 1338 (7)
13
Study Objectives

• Describe mortality risk by demo-graphic and
  clinical characteristics
• Compare cause-specific mortality rates with U.S.
  population
• Assess treatment factors associated with excess
  cause-specific mortality

14
Methods

• National Death Index search complete as of
  12/31/2002
• Review of death certificates obtained on U.S.
  cases
• Cause of death coded using ICD-9
• Categorization of cause of death

15
Underlying cause of death

• Direct consequence of original cancer
• Treatment-related sequelae
• Subsequent malignant neoplasm
• Cardiac-related
• Pulmonary
• Non-treatment related
• Accidents
• Other causes

16
Deaths

• Total deaths 2823
• - Death certificate received 2435
  (84)
• - Death cause, other sources 121 (4)
• - Unable to determine cause 267 (12)

17
All cause mortality- sex specific survival
18
Overall mortality

• deaths SMR
  (95 CI)
• Total 2823 8.23 (7.9-8.5)
• Sex
• Male 1687 6.6 (6.3-6.9)
• Female 1136 13.1 (12.3-13.9)
• Years survived after diagnosis
• 5-9 1336 21.4 (19.3-22.6)
• 10-14 611 7.1 (6.5-7.7)
• 15-19 431 4.6 (4.2-5.0)
• 20-24 269 4.3 (3.8-4.8)
• 25-29 145 4.9 (4.1-9.6)

19
Overall mortality by diagnosis

• deaths SMR
  (95 CI)
• Leukemia 913 9.9 (9.3-10.6)
• CNS Tumors 547 12.5 (11.4-13.5)
• Hodgkins Disease 510 7.7 (7.0-8.4)
• Non-Hodgkins 143 4.4 (3.7-5.2)
• Kidney (Wilms) 98 4.6 (3.7-5.6)
• Neuroblastoma 84 5.5 (4.3-6.7)
• Soft Tissue Sarcoma 244 7.0 (6.2-8.0)
• Bone Tumors 284 7.7 (6.9-8.7)

20
Causes of death

• Recurrence 1469 57
• Treatment-related 703 28
• Other causes 384 15

21
Death due to recurrence by original diagnosis
22
Death due to recurrence by original diagnosis
23
Recurrence related deaths

• deaths
  Rate (/year) (95 CI)
• Sex
• Male 882 0.47 (0.44-0.50)
• Female 587 0.38 (0.35-0.41)
• Years Survived after Diagnosis
• 5-9 978 0.98 (0.93-1.05)
• 10-14 307 0.32 (0.29-0.36)
• 15-19 116 0.14 (0.12-0.17)
• 20-24 48 0.10 (0.08-0.14)
• 25-29 19 0.10 (0.06-0.16)

24
Cumulative cause specific mortality
25
Treatment-related mortality

• deaths SMR (95
  CI)
• New cancer 380 15.0
  (13.7-16.4)
• Cardiac 179 6.9
  (5.8 -8.1)
• Pulmonary 45 8.7 (6.7
  -11.0)

26
Subsequent cancer mortality

• Relative risk (95 CI)
  p-value
• Radiation 2.3 (1.7-3.0)
• Alkylating agents 1.3 (1.0-1.6) 0.03
• Anthracyclines 1.0 (0.8-1.3) 0.73
• Epipodophyllotoxin 1.7 (1.2-2.4)
• Bleomycin 1.3 (0.9-1.9) 0.14
• Adjusted gender, diagnosis age, death age,
  years since diagnosis

27
Cardiac mortality

• Relative risk
  (95 CI) p-value
• Radiation to heart 2.4 (1.7-3.4)
• Alkylating agents 0.9 (0.6-1.4) 0.64
• Anthracyclines 2.1 (1.3-3.3)
• Epipodophyllotoxin 0.8 (0.3-1.8) 0.56
• Bleomycin 1.5 (0.9-2.7)
  0.11 Adjusted gender, diagnosis age,
  death age, years since diagnosis

28
Pulmonary mortality

• Relative risk
  (95 CI) p-value
• Radiation to chest 1.2 (0.7-2.1) 0.52
• Alkylating agents 1.1 (0.6-2.0) 0.76
• Anthracyclines 1.0 (0.5-2.0) 0.94
• Epipodophyllotoxin 1.7 (0.6-4.5) 0.29
• Bleomycin 3.6 (1.1-12.7) 0.05
• Adjusted gender, diagnosis age, death
  age, years since diagnosis
  

29
Non-treatment related mortality

• deaths SMR
  (95 CI)
• External causes 182 0.9 (0.8
  -1.0)
• auto accident 80 1.0
  (0.8-1.3)
• other accidents 49 1.3
  (1.0-1.7)
• suicide 39
  1.0 (0.7-1.4)
• homicide 23 0.5
  (0.3-0.7)

30
Conclusions

• The pattern of deaths in five-year survivors is
  key to understanding late relapse and the late
  effects of treatment.
• Overall, late mortality is lower in patients
  treated in more recent time periods.
• Exposure to specific chemotherapies and radiation
  are associated with cause-specific mortality risk

31
Conclusions (cont.)

• In cohorts with extended follow-up, overall death
  rate is still 8.2 times that expected in general
  population
• Recurrence is the main cause of death, but rate
  decreases with time since diagnosis
• For treatment related deaths, rates continue to
  increase with time since diagnosis

32
Future Direction

• Long-term childhood cancer survivors need
  continued follow-up for emerging patterns of
  late-occurring mortality
• Are there external factors that influence
  increased mortality rate (i.e., health behaviors,
  early screening)?
• What is the role of genetics in overall survival?

33
Recurrence related deaths

• deaths Rate
  (/year) (95 CI)
• Diagnosis
• Leukemia 593 0.55 (0.50-0.59)
• CNS Tumors 335 0.75 (0.67-0.83)
• Hodgkins Disease 154 0.32 (0.27-0.38)
• Non-Hodgkins 38 0.15 (0.10-0.20)
• Kidney (Wilms) 22 0.07 (0.04-0.11)
• Neuroblastoma 43 0.18 (0.13-0.24)
• Soft Tissue Sarcoma 120 0.37 (0.31-0.45)
• Bone Tumors 164 0.58 (0.49-0.67)