1 The Application of The Application of Evidence-Based Medicine Evidence-Based Medicine to Achieve Progress in to Achieve Progress in Pediatric Oncology Pediatric Oncology Malcolm Smith, MD, PhD Malcolm Smith, MD, PhD Pediatric Section, Clinical Pediatric Section, Clinical Investigations Branch Investigations Branch 12 September, 2000 12 September, 2000 http://ctep.info.nih.gov/
49
Embed
1 The Application of Evidence-Based Medicine to Achieve Progress in Pediatric Oncology Malcolm Smith, MD, PhD Pediatric Section, Clinical Investigations.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
11
The Application of Evidence-Based The Application of Evidence-Based Medicine to Achieve Progress in Medicine to Achieve Progress in Pediatric OncologyPediatric Oncology
Introduction and historical perspective Importance of phase III randomized
clinical trials to progress Importance of risk-adjusted therapy to
developing better treatment strategies Clinical trials research infrastructure Unmet needs and future directions
Childhood CancerChildhood Cancer
8700 new cases diagnosed annually in children younger than 15 years of age and 12,400 cases in persons younger than 20 years
Approximately 1700 children younger than 15 years and 2300 children/adolescents younger than 20 years die each year in U.S.
Most of the cancers of children differ from those of adults in their histology and in their biological characteristics
Age-Adjusted SEER Cancer Incidence, 1991-95, Age-Adjusted SEER Cancer Incidence, 1991-95, Reflecting Cancers Occurring among AdultsReflecting Cancers Occurring among Adults
Prostate17.1%
Breast14.7%
Lung14.2%
Colorectal11.0%
Bladder4.1% NHL
3.9%Uterus2.8%
Other32.2%
Distribution of Cancer Diagnoses: 0-14 Years
ALL23.5%
AML4.7%
Brain22.1%
Neuroblastoma7.9%
Wilms' tumor6.0%
NHL5.7%
Hodgkin's3.6%
Rhabdomyosarcoma3.6%
Non-RMS3.5%
Germ cell (gonadal)3.5%
Retinoblastoma3.2%
Osteosarcoma2.6%
Other 10.1%
Childhood Cancer Clinical ResearchChildhood Cancer Clinical Research
National efforts are essential for studying specific childhood cancers because of the limited numbers of children with individual cancer types
The NCI has supported since the 1950s a nationwide clinical research program specifically designed to improve the outcome and quality of life for children with cancer
Justification for Separate Studies for Justification for Separate Studies for Children with CancerChildren with Cancer
The cancers of children are generally biologically distinctive from those occurring in adults.• The response of childhood cancers to anti-cancer
treatments may be qualitatively or quantitatively different from that of adult cancers
The ability of children to tolerate anti-cancer treatments may differ from that of adults.
Investigators with special expertise in pediatric oncology are best qualified to prioritize, design, and implement clinical trials for children with cancer.
Survival Rates Have Dramatically Survival Rates Have Dramatically Improved for Children with CancerImproved for Children with Cancer
Small minority of children cured of their cancer in 1960
Current 5-year survival rates for children with cancer < 15 years = 75%
Childhood cancer mortality rate decreased nearly 50% from 1973-96
The decrease in childhood cancer mortality continued in the 1990s at rate of 2.7% per year
SEER Cancer Statistics Review, 1973-96http://www-seer.ims.nci.nih.gov/Publications/
Remarkable Past Progress Leukemia Mortality 1950-96
Remarkable Past Progress NHL Survival Rates:1960-1996Survival Rates:1960-1996
18%
45%
62%
78%
0%
20%
40%
60%
80%
100%
1960-63 1974-76 1980-82 1989--96
Age < 15 Years SEER (9 areas)
http://www-seer.ims.nci.nih.gov/
Contributions of NCI-Supported Nationwide Clinical Trials System to Improved Outcome
Conducting randomized phase III clinical trials that reliably identify superior new treatments
Providing children with cancer throughout the United States and Canada access to state-of-the-art treatment protocols that are developed by national experts and that have multiple levels of review for scientific quality and multiple levels of review for patient safety
Providing central review of pathology and imaging leading to improved diagnosis and staging
Supporting research studies leading to the identification of reliable clinical and biological prognostic factors
Why Randomized Phase III Why Randomized Phase III Clinical Trials??Clinical Trials??
Because what is logical and should work often doesn’t:• Identifying new superior treatments is an empirical,
not a deductive process
Example: Anti-arrhythmic therapy to prevent mortality from fatal arrhythmias
1. Elevated VPBs are associated with early death
2. Encainide and flecainide suppress VPBs
3. Therefore, encainide and flecainide should reduce mortality in patients with VPBs
WRONG!
Why Randomized Phase III Why Randomized Phase III Clinical Trials??Clinical Trials??
Need reliable answers to questions of therapy:• Example: Accepting a more toxic therapy as superior when
it is not better than standard therapy has serious consequences for future patients.
Conclusions from single arm (non-randomized) clinical trials have limited reliability:• Apparent improvements ascribed to treatment are often
due to patient selection (selection bias)– Historical control populations may differ from current study
populations
• Improvement ascribed to one intervention may be due to second uncontrolled factor (supportive care, XRT, surgery).
Selection Bias in High-dose Selection Bias in High-dose Chemotherapy (HDCT) TrialsChemotherapy (HDCT) Trials
Single arm trials suggested higher response rates and survival rates for HDCT in women with breast cancer
Outcome for 1581 patients with metastatic breast cancer treated with conventional doxorubicin-based regimens, not HDCT: (J Clin Oncol 15:3171,1997)
CR% MedianPFS
5-YrSurvival
HDCTEligible
27% 16 mos 21%
HDCTIneligible
7% 8 mos 6%
NCI-Supported Clinical Research for NCI-Supported Clinical Research for Children with Cancer--Phase III TrialsChildren with Cancer--Phase III Trials
Phase III trials generally require 100s of patients to reliably identify clinically meaningful differences between treatments being compared
Patients randomized to receive “best available” therapy or to receive a new treatment• The new treatment is prioritized for evaluation based on
preliminary data suggesting its potential for improving outcome (i.e., increase survival, diminished toxicity)
Address important questions of therapy for which the answer is not known
NCI-Supported Clinical Research for NCI-Supported Clinical Research for Children with Cancer--Phase III TrialsChildren with Cancer--Phase III Trials
Participation in Phase III trials is considered an appropriate “standard of care” for children with cancer:• Rationale: Many current treatments are sub-optimal because
of limited efficacy and/or excessive toxicities• Safeguards for patient protection: multiple levels of scientific
review and review for patient safety, and appropriate informed consent/assent
• Given the above, it is felt appropriate in most circumstances to ask families to consider participation in phase III trials
Phase III trials available for most types of childhood cancer • ~ 25 phase III trials at any given time
Randomized Trials for Pediatric Acute Randomized Trials for Pediatric Acute Lymphoblastic Leukemia: CCG-1922Lymphoblastic Leukemia: CCG-1922
S ta nd a r d R isk A L L
P red nisoneO ra l 6M P
O P
P red nisoneIV 6M P
IP
D exam ethasoneO ra l 6M P
O D
D exam ethasoneIV 6M P
ID
N ew ly D iagnosed A L LS tand ard R isk
CCG-1922 EFS FROM ON STUDY (RANDOMIZATION) BY REGIMEN
0.5
0.6
0.7
0.8
0.9
1
0 1 2 3 4 5 6
YEARS
PR
OB
AB
ILIT
Y
OP
IP
OD
ID
(N=270)
(N=260)
(N=274)
(N=256)
From Phase I to Phase From Phase I to Phase III--Ifosfamide/Etoposide for Ewing’s III--Ifosfamide/Etoposide for Ewing’s SarcomaSarcoma
1986: Ifosfamide reported as active agent for Ewing’s sarcoma
1987: Ifosfamide/etoposide (IE) combination reported as active for Ewing’s sarcoma
1988: Phase III trial evaluating IE for Ewing’s sarcoma begins
1994: Phase III trial closes 1995: Phase III trial results reported: IE
improves outcome for Ewing’s sarcoma
Randomized Trials for Pediatric Randomized Trials for Pediatric Cancers - Ewing’s SarcomaCancers - Ewing’s Sarcoma
IN T -0 0 9 1 - In te r g r o up E w ing ’s S a r c o m a S tud y
V cr + D ox + C ycT reat q 3 w eeks
V cr + D ox + C ycA ltern atin g w ithIf osf am id e + E top osid eT reat q 3 w eeks
L oca lized E w in g 's S arcom a
Randomized Trials for Pediatric Randomized Trials for Pediatric Cancers - Ewing’s SarcomaCancers - Ewing’s Sarcoma
Trial Results: Patients receiving the ifosfamide/etoposide combination (n=198) had superior 3-yr EFS rates compared to pts receiving standard therapy (n=200): • 69% 3-yr EFS vs 50% 3-yr EFS (p = 0.0005)
Significance: The INT-0091 defined a new standard therapy for Ewing’s sarcoma that includes ifosfamide + etoposide.
Determining this required a commitment of resources for over a decade from the time of the initial evaluation of ifosfamide in children
Results from Recent Phase III Trials for Results from Recent Phase III Trials for Children with CancerChildren with Cancer
ABMT is superior to intensive conventional therapy as consolidation therapy for children with neuroblastoma
Cis-retinoic acid maintenance therapy following ABMT improves outcome for children with neuroblastoma
Reducing the dose of craniospinal radiation for children with medullblastoma increases the relapse rate
Pulse-intensive short courses of therapy have similar efficacy as more burdensome, long courses of therapy for children with Wilms’ tumor
Study Questions for Recently Study Questions for Recently Completed/Ongoing Phase III TrialsCompleted/Ongoing Phase III Trials
High-dose methotrexate for T-cell ALL Dexrazoxane as a cardioprotectant for children with T-cell
ALL and children with Hodgkin’s disease Dose-intensive therapy for Ewing’s sarcoma Ifosfamide and MTP-PE for osteosarcoma Triple intrathecal therapy for children with ALL Defining optimal thiopurine for children with ALL MDR-reversal agent for children with AML Ch14.18 (anti-GD2 monoclonal antibody) for children with
neuroblastoma following ASCT Topotecan + cyclophosphamide for rhabdomyosarcoma
CCG: Survival of Children withAcute Lymphoblastic Leukemia
8080
6060
4040
2020
100100 Years of Years of Number of Number of
1972-751972-751972-751972-75
00 22 44 8866 1010
Years from DiagnosisYears from Diagnosis
%%
SSuurrvviivvaall
1989-931989-93
1983-891983-891978-831978-83
1970-721970-72
1968-701968-70
3,0803,080
3,7123,7122,9792,9791,3131,313
936936
499499
402402
DIagnosisDIagnosis ChildrenChildren
12,92112,921Total Number ofTotal Number ofPatients Treated:Patients Treated:
C C GC C GBleyerBleyer
Risk-Adjusted Therapy:Risk-Adjusted Therapy:Classifying Patients by PrognosisClassifying Patients by Prognosis
The approach to treatment (and to clinical trial design) differs based on the anticipated outcome for the population.
Patients destined to survive
Patients destined to have poor outcome
Risk-Adjusted Therapy:Risk-Adjusted Therapy:Classifying Patients by PrognosisClassifying Patients by Prognosis
If prognostic factors can be identified that allow identification of which patients do well with current therapy and which do poorly, this allows treatment intensity/risk to be better tailored to likely outcome.
Patients destined to survive
Patients destined to have poor outcome
Poor Prognosis Group
Favorable Prognosis Group
Risk-Adjusted Therapy: Assigning Risk-Adjusted Therapy: Assigning Treatment Based on PrognosisTreatment Based on Prognosis
Patients who have low survival rates with current treatments may benefit from novel, more aggressive therapeutic approaches that are associated with greater risk.
Patients who have very good outcome with current therapy should be spared more intensive and toxic treatments.
Risk-Adjusted Therapy Requires the Risk-Adjusted Therapy Requires the Identification of Reliable Prognostic FactorsIdentification of Reliable Prognostic Factors
Identification of reliable prognostic factors:• Requires analyzing outcome for large
numbers of patients, preferably treated in uniform manner.
• For biological prognostic factors, requires collection and analysis of tumor tissue.
Protocol-treated patients and Cooperative Group tumor banks have been invaluable in identifying and confirming prognostic factors
Scope of NCI-Supported Pediatric Scope of NCI-Supported Pediatric Clinical Trials ProgramClinical Trials Program
Approximately 5000 children entered onto treatment trials each year.• 3,800 Phase III entries to ~ 25 trials• 700 Phase II entries to ~ 30 trials• 250 Phase I to ~ 25 trials
Accrual to Phase III by tumor type:• ALL ~ 1,900 (2,300)• AML ~ 410• Wilms ~ 480• Ewing’s ~ 160• Osteosarcoma ~230
Cooperative Groups supported via Cooperative Agreements with the NCI:• Children’s Cancer Group• Pediatric Oncology Group• Intergroup Rhabdomyosarcoma Study Group• National Wilms’ Tumor Study Group
Represent over 200 institutions throughout U.S. and Canada that are involved in the treatment of most children with cancer
Children’s Oncology Group Merger of pediatric clinical trials groups into
single entity: the Children’s Oncology Group• Improved efficiency in developing and conducting
clinical trials for children with cancer.
Pediatric Cooperative Group Clinical Trials Pediatric Cooperative Group Clinical Trials Program --Participating ResearchersProgram --Participating Researchers
Pediatric Cooperative Group Clinical Pediatric Cooperative Group Clinical Trials Program --Supported Structures 1Trials Program --Supported Structures 1
Operations Office• Coordinate protocol development & distribution• Organize semi-annual meetings• Distribute funds to member institutions• Regulatory oversight & negotiate contracts with pharmaceutical
companies when appropriate
Statistical Center• Statistical design of protocols• Data collection and management• Assure IRB review prior to protocol entry• Analysis of data from trials• Institutional performance review• Coordinate on-site audit program
Pediatric Cooperative Group Clinical Pediatric Cooperative Group Clinical Trials Program --Supported Structures 2Trials Program --Supported Structures 2
Member Institutions• Principal investigator at institution• Clinical research associates• Partial reimbursement for research costs of
patient accrual (~$1,725 per patient direct costs)
• Support for tissue collection and shipping to central reference laboratory/tumor bank
• Support for submitting radiographs, pathology reports, surgical reports, etc.
Pediatric Cooperative Group Clinical Trials Pediatric Cooperative Group Clinical Trials Program --Supported Structures 3Program --Supported Structures 3
Disease and Discipline Committees• Committees responsible for developing the
questions of therapy for clinical trials• Examples of Disease Committees: ALL, AML,
Neuroblastoma, Bone Tumor, etc.• Examples of Discipline Committees: Surgery,
Radiation Oncology, Nursing, etc.• Study Committees for developing and
implementing individual clinical trials generally assigned by Disease Committees.
Survivors of Childhood CancerSurvivors of Childhood Cancer
At risk for long-term sequelae of therapy and for sequelae of cancer itself:• Cardiac and Pulmonary• Second Neoplasms• Fertility and Offspring• Central Nervous System• Musculoskeletal• Psychosocial
Retrospective Cohort n ~ 13,000Newly Diagnosed 1970-1986<21 yrs at DiagnosisSurvived 5 yrs. from DiagnosisEnglish-, Spanish-speakingReside in U.S. or Canada
Surveyed for long-term health and psychosocial status
Childhood Cancer Survivor Childhood Cancer Survivor Study (CCSS): - Study DesignStudy (CCSS): - Study Design
Selected CCSS Analyses in ProgressSelected CCSS Analyses in Progress
Late mortality in childhood cancer survivors Second malignant neoplasms following
childhood cancer Pregnancy outcomes after treatment for
cancer during childhood or adolescence Cancer in offspring of pediatric cancer
patients Thyroid disease in survivors of childhood
and adolescent Hodgkins disease Smoking among childhood cancer survivors
Unmet NeedsUnmet Needs Over 2000 children and adolescents die
from cancer each year in U.S. Some children who are cured experience
diminished quality of life because of the long-term effects of their cancer diagnosis and treatment
Current therapy is near-maximal intensity, and new treatment strategies are needed to improve outcome for these children
Figure 11
Leukemia 33.0%
CNS 24.0%
Endocrine 9.0%
Hodgkin's 1.0%
NHL 6.0%
Kidney 3.0%
Liver 2.0%
Bone 7.0%
Soft tissue 7.0%
Other 8.0%Distribution of Cancer Mortality in Children 0-19 Years
Leukemia 33%
Moving Towards a New EraMoving Towards a New Era Molecularly targeted therapies:
Treatments based on the specific molecular characteristics of the cancer
In principle, more specific for processes required for tumor cell survival and growth
Bcr-Abl as Target for Treatment of Bcr-Abl as Target for Treatment of Philadelphia Chromosome Positive ALLPhiladelphia Chromosome Positive ALL
Ph+ ALL with the Bcr-Abl fusion protein has very poor outcome among children
Bcr-Abl fusion protein has an enzyme activity (tyrosine kinase) necessary for leukemogenic effect.
STI571: Inhibitor of the Bcr-Abl, PDGF, and c-KIT receptor protein-tyrosine kinases. • Inhibits proliferation & induces apoptosis
Nat Med 2:561, 1996 & Cancer Res 56:100, 1996Blood 90:4947,1997 & Clin Can Res 4:1661,1998
Mechanismof Action
9;22 translocation
bcr-abl fusion protein
Normal hematopoiesis
Ph+ ALL
STI571
bcr- abl fusion protein
Bcr-Abl as Target for Bcr-Abl as Target for CML and PhCML and Ph++ ALL Therapy ALL Therapy
Phase I trials completed in adults with CML with very high levels of anti-leukemia activity observed
Pediatric phase I trial ongoing Pilot study for newly diagnosed
patients with Ph+ ALL planned for early 2001
Summary 1 The public health of children has been
improved by long-term, sustained NIH support of an ongoing infrastructure for conducting clinical research for children with cancer.
Superior new treatments have been identified based on definitive evidence, and these treatments have been made widely available to children with cancer throughout the United States and Canada.
Summary 2: Progress Depends on Collaboration and Cooperation
Pediatric cancer researchers (clinical and laboratory) and health care professionals
Families and their advocates National Cancer Institute Academic and pharmaceutical developers of
new cancer treatments and the FDA Third party payers Working together so that the most promising
therapeutic approaches are expeditiously evaluated with the ultimate objective of improving outcome for children with cancer