paediatrics Brussels 17

Epidemiology of Cancer in Early Life

David A Walker University of Nottingham www.cbtrc.uk www.headsmart.org.uk

Objectives

• To debate health priority for cancer services in early life • To consider the impact of age growth and development upon risk of cancers in early life. • To explore the impact of normal processes of adolescence upon an individual’s capacity to participate in treatments and trials. • To propose strategies for enhancing outcome within clinical trials for TYA population

All Cancers Excluding Non-Melanoma Skin Cancer (C00-97 Excl. C44): 2008-2010 Average Number of New Cases Per Year and Age- Specific Incidence Rates per 100,000 Population, UK

Debate :

Specialist cancer services for children and young people are a vital health priority

Arguments for Arguments against

2 minutes

Objectives

• To consider the impact of age growth and development upon risk of cancers in early life.

All Childhood Cancers: 1988-1997World Age- Standardised Incidence Rates per Million Population, Children (0-14), Europe

All Childhood Cancers: 2006-2007Average Number of New Cases Diagnosed per Year, Children (0-14), Great Britain

All Childhood Cancers*: 1966-1970 to 2006-2007 World Age-Standardised Incidence Rates per Million Population, Children Aged 0-14, Great Britain

Cumulative cancer risk:

It has been estimated that around one child in every 500 will develop some form of cancer by 14 years of age in Great Britain

1 in 200 by age 22 years

Childhood malignancies: Cytology / Imaging

Normal Abnormal

Normal

Leukaemia

Osteosarcoma

Cerebellar astrocytoma

Wilms Tumour

Metastases

Lymphoma

mIBG Scan

Neuroblastoma

Evidence for a Developmental Hypothesis for Childhood Cancer Scotting Perilongo & Walker Nature Reviews Cancer (2005) 5; 481-488

• Fidelity of Embryonic / Fetal / Immature

Cell Tissue

microenvironment

• Teratocarcinomas

Mimic many tissues. If transplanted back into growing embryos revert to normal cell behaviour (Martin 1981, Andrews 2002)

A Developmental Hypothesis for Childhood Cancer Scotting Perilongo & Walker Nature Reviews Cancer (2005) 5; 481-488

Children’s Tumours • Differ in their frequency and tissues of origin, • numbers and types of genetic mutations, • sensitivity to chemotherapy •

arise within growing and developing organs, at a time when the tissue microenvironment promotes rapid growth development. The majority are sporadic. Although the more you look, the more predisposing mutations you find Normal developing tissues • have biological characteristics that sustain growth and development: ie. sustained cell division, migration and resistance to cell death • Tissue growth rates decelerate after birth through to end of adolescence.

Clinical phenomena where tumour growth arrests or involutes • Congenital Haemangioma * • Congenital Cardiac Rhabdomyoma of Tuberous Sclerosis by 1-2 years • Arrest of TAM in Downs by 6/12 - 1 year • Neuroblastoma 4s involution by 6/12 – 1 yr • Infantile fibrosarcoma, by 1 yr • Arrested progression of infantile multi-system Langerhans Cell Histiocytosis by 3-5 yrs • Arrested development / progression retinoblastoma by 5-6 yrs • Arrested tumour growth pilocytic astrocytoma *

Bilateral retinoblastoma

LCH

4 m

16m

28m

42m

Capillary Haemangioma

Haemangioma

Bilateral retinoblastoma

LCH

Leukaemias: 1996-2005 Incidence Rates per Million Population, Children (0-14), Great Britain

Why do so many children get ALL

A higher risk of ALL in children living in more affluent areas, compared with the most deprived, has persisted in Great Britain over many decades, including in the latest analysis, with data on all children diagnosed in England & Wales up to 2005

Hypotheses for aetiology of ALL

• The ‘delayed infection’ hypothesis some childhood leukaemia are the result of a rare response to an unidentified infection following geographic or social isolation early in life);27 • The ‘population mixing’ hypothesis leukaemia results from an unfamiliar or uncommon infection which the child is exposed to through new contact with people from different geographical areas

Embryonal Tumours: 1996-2005 Incidence Rates per Million Population, Children (0-14), Great Britain

Now account for < 30% of Wilms Tumour WT1 Frasier Syndrome Denys Drash WAGR p53

Kidney Cancer (C64-C66,C68): 2008-2010 Average Number of New Cases Per Year and Age- Specific Incidence Rates per 100,000 Population, UK

Liver Cancer (C22): 2008-2010 Average Number of New Cases Per Year and Age- Specific Incidence Rates per 100,000 Population, UK

Low Med High

L1

M

L2

Bone and Soft Tissue Sarcomas: 1996-2005 Incidence Rates per Million Population, Children (0-14), Great Britain

Hodgkin Lymphoma (C81): 2008-2010 Average Number of New Cases Per Year and Age- Specific Incidence Rates per 100,000 Population, UK

Brain CNS and other intracranial tumours Age-Specific Incidence Rates per 100,000 Population, UK

Gonadal and Germ Cell Tumours: 1996-2005 Incidence Rates per Million Population, Children (0-14), Great Britain

Testicular and Ovarian

Bone and Connective Tissue (C40-C41,C47,C49): 2006- 2008 Average Number of New Cases Per Year and Age- Specific Incidence Rates per 100,000 Population, UK

Ewings

Osteosarcoma

Suggest overarching hypothesis for cause of cancer in early life

Objectives

• To explore the impact of normal processes of adolescence upon an individual’s capacity to participate in treatments and trials.

Changing relation b/t puberty & psychosocial transitions into adulthood Patton & Viner, The Lancet March 2007

age

menarche

30

Psychosocial maturation

Mismatch biological & psychological transitions

20

10

Mid 20 C

HG

Agric.

IR

0

2000

20,000

200 50 now

Gray matter density changes 5yrs to 20yrs

Developmental stage

Am I normal? Early

(10-14)

biological focus

Confidentiality

Who am I?

Middle (15-17)

peer focus

Where am I going? Late (18+)

educational/vocational intimate relationships

Communication style implications

At the start of the consultation...

Explain terms & exceptions because...

Ford et al . JAMA, 1997

* P<0.001

A time of immense change

3 Clocks

Emotional

Intellectual

Physical

Development in each can be at different rates and do not necessarily sync with chronological age

Slide courtesy of Dr Michael Carr-Gregg

Risk and protective factors

risk

protective

School World

risk

risk

protective

protective

Family World

Peer World

risk

protective

Inner World

Slide CourtesyDr Michael Carr-Gregg

Interregional correlations of fMRI measurements of Latent Variable Interval between paired brain regions in subjects identified by testing to have high or low ResistanceTo Peer Influence (RPI) and exposed to films of aggressive hand waving.

Challenges that Face 15-30 Year-Olds

HIV, STDs

Pregnancy

Stamina, Endurance

Fertility

Significant Other

Marriage, Spouse Conflicts

Children, Parenthood

Domicile, Home

Alcohol, Drugs, Nicotine, Addiction

Cancer, Suicide: The Most Common Disease Killers

Autonomy, Independence

Maturity

Education

Athleticism

IQ, EQ

Peer Pressure, Social Acceptance

Career Choice

Employment

Sexuality

Health Insurance

Self-Image, Body Image

Growth

• H ome and Relationships • E ducation and Employment • E ating • A ctivities and Hobbies AT THIS STAGE REASSURE ABOUT CONFIDENTIALITY • D rugs, Alcohol and Tobacco • S ex and Relationships • S elf harm and Depression • S afety and Abuse

The teenager is your patient!

“Anything we talk about today is confidential. That means I cannot tell others, including your parents, about it without your permission. The only exceptions would be if I thought you, or someone else, was at risk of serious harm. In that case I would need to tell someone else.”

But don’t ignore the parents!

• HOME: How is it at home at the moment? Do you have your own space? Who do you get on best with? Could you talk to them if you were worried about anything? • EDUCATION: How’s school going? What are you best at? Do you know what you want to do when you leave? Do you have a good group of friends? • EATING: Has your weight changed recently? Are you worried about it? Have you ever dieted? How much exercise do you get? • ACTIVITIES AND HOBBIES: Do you have a good social life? What do you do to relax? • DRUGS, ALCOHOL, TOBACCO: Lots of teenagers try smoking/alcohol, have you? Have you been offered drugs? Is it hard for you to say no in this situation? • SEX AND RELATIONSHIPS: Young people can start to develop intimate relationships, have you handled that part of your relationship yet? What do you know about contraception? Have you ever felt pressured? • SELF HARM AND DEPRESSION: How is life in general? How are you sleeping? Do you ever think about hurting yourself? Do you ever feel so down that life isn’t worth living? • SAFETY AND ABUSE: have you ever been seriously injured? Have you ever been in a fight? Is anyone harming you, or making you do things you don’t want to?

TYA Cancer Statistics • 173,000 cases of cancer diagnosed TYA worldwide in 2008 • > 3-fold variation in world incidence rates between regions • In European Union, ~14,700 cases of TYA cancer in 2008. • Age-specific incidence rate for all cancers in 15-19 year-olds in 19 European countries increased from 147 per million (1970s) to 165 (1980s) and 193 (1990s) • Ranked increases: carcinomas (3.9%), soft tissue sarcomas (2.6%), lymphomas (2.4%), GCTs (1.7%) and CNS tumours (1.4%). • US age-specific incidence rate was 220 per million (15-19yrs) and 371 (20–24 years) • US incidence rates have increased significantly in both age- groups since 1975: Except STS and Carcinoma (15-19)

Health and Educational Services During TYA period

Consent

Legal E&W Sexual Legal Scot Primary Care Primary Care Primary Care

Specialty adolescent health services

Hospital / community paediatric services

Hospital / community adult services

Involvement with assent

University /

College

Seconday Education

Primary Education

Vocational

Boys pubertal development

Girls pubertal development

9 10 11 12 13 14 15 16 17 18 19 20 21 22

TYA: Consent, Rights, Independence and Health Care

Consent

Legal E&W Sexual Legal Scot Primary Care Primary Care Primary Care

Specialty adolescent health services

Hospital / community paediatric services

Hospital / community adult services

Financial Independence

Parental dependence

Best interests of child / TYA

Adult Human Rights

Involvement with assent

Boys pubertal development

Girls pubertal development

…9 10 11 12 13 14 15 16 17 18 19 20 21 22

Objectives

• To propose strategies for enhancing outcome within clinical trials for TYA population

Cancers in Older Adolescents and Young Adults SEER, 1975-1998

Age 15-19 Years

Testis 9% Ovary 7%

Other 12%

Thyroid

7%

Germ Cell

Melanoma 7%

Hodgkin's 16%

Lymphoma

Sarcoma

Ewing's 2%

Brain

Leukemia

Osteosarcoma 5%

Age 20-29 Years

NHL

8%

ALL

AML 5%

CNS

Soft Tissue Sa

6%

Other Carcinomas 10%

10%

Other 2%

7%

Germ-Cell 17%

GI Carcinoma 4%

Breast Carcinoma 5%

Lymphomas

18%

Thyroid Carcinoma 10%

The segments in color represent “paediatric malignancies”

Soft-tissue

Malignant Melanoma 12%

Bone

Sarcomas

Leukemia 5%

CNS 6%

10%

National Treatment Trial Accruals, 1990-1998 National Cancer Mortality Reduction, 1990-1998

25%

20%

% Mortality Reduction

12,000

15%

Accruals

3%

10%

Cancer Mortality Reduction

5%

8,000

p = .001

0%

2%

1,000

10,000

Accruals

4,000

1%

0

0%

0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 Age (Years)

Buzz Group

If survival rates are poorer for this TYA age group: what strategies within clinical trials could be adopted to enhance outcomes?

European Survival Statistics Childhood Cancer

Survivorship: UK statistics 2005-2012

• At the end of 2005, it was estimated that around 26,000 people in Great Britain were long-term childhood cancer survivors, who had survived five years or longer after diagnosis with childhood cancer. • It is estimated that by the end of 2012 there will be at least 33,000 people in the UK who are alive having previously been diagnosed with a childhood cancer and who survived their cancer for at least five years.

Our Objectives?

• To debate health priority for cancer services in early life • To consider the impact of age growth and development upon risk of cancers in early life. • To explore the impact of normal processes of adolescence upon an individual’s capacity to participate in treatments and trials. • To propose strategies for enhancing outcome within clinical trials for TYA population

Imaging in Medulloblastoma and Ependymoma

Tim Jaspan University Hospital Nottingham

MEDULLOBLASTOMA

Medulloblastoma - demographics

• Commonest pediatric malignant brain tumor • 15-20% of all pediatric brain tumors • 80% <15 years old, 20% <2 years old • 40% of pediatric posterior fossa tumors • Slightly more common in boys • Mean age at diagnosis – 7yrs; peaks at 1-5 yrs and 6-9yrs

Medulloblastomas

1-5 yr: • Often arise from inferior medullary velum of the vermis • Usually presents as a midline tumor • Spherical pattern of growth, projecting into enlarged 4V • Hydrocephalus at diagnosis in 95% of case

6-9 yr + older: • Sometimes mimics cerebellopontine angle tumor • More often in cerebellar hemispheres

Molecular classification

• New molecular subgroups identified • SHH, WNT, Group 3, Group 4

• Subgroups show different clinical/biological behaviour • Potential for improved risk stratification/tailored treatment

Subgroup features

• WNT : occur mainly along the CP/CPA axis - good prognosis • SHH : >50% cerebellar hemispheric - intermediate prognosis • GP 3 : primarily midline/juxtaventricular, enhancement +/++ - poor prognosis, early dissemination • Gp 4 : midline/4 th ventricle based, enhancement 0/+ - dissemination common

Imaging

Diagnostic modalities: • CT - Restricted to initial diagnosis • MRI - Neuraxial MRI is the standard of care

Medulloblastoma - CT

• Often first diagnostic imaging • Mildly hyperdense (unusual for post. fossa tumors); reflects high cellular density/nuclear-to-cytoplasmic ratio • Calcifications in approx 20% • Cysts in 50-60%

• Surrounding oedema often seen • Enhancement variable/prominent

CT

Medulloblastoma - MRI

• Iso to mildly T1 hypointense/mixed T2 SI • Grows circumferentially to fill/obliterate 4V • Doesn’t usually extend out through foramen of Luschka • More aggressive lesions invade brainstem and/or adjacent cerebellar parenchyma • Haemorrhage, necrosis or cystic change in 20-50% • Generally strong or heterogeneous enhancement • DWI: typically restricted diffusion (dark on ADC maps) • MRS: elevated Choline and Taurine, low NAA

MB - WNT

T2

T1

T1 post Gd

T1 post Gd

MB - SHH

T1

T1 post Gd

T1 post Gd

T2

DWI

ADC

MB - Gp 3

T2

T1

T1 post Gd

T1 post Gd

DWI

ADC

MB - Gp 4

T2

FLAIR

CISS

DWI

T1 post Gd

ADC

T1

T1 post Gd

MR Spectroscopy – short echo

Differential diagnosis

ATRT

Ependymoma

JPA

Distant spread

• Must image entire neuraxis • Must image before surgery • Spinal imaging < 3 weeks from surgery inaccurate (false positives) – can lead to delay in definitive surgery • Metastatic spread in approx 20% at diagnosis • Local CSF dissemination appears more linear • Distant spread often more nodular, but also linear • 5% extra CNS spread

Metastatic spread at diagnosis

Oct ‘13 Post i/t etopiside

Sept ‘12 Pre op

Leptomeningeal spread

Pre Gd FLAIR

Post Gd FLAIR

Post Gd T1

Nodular spinal disease

Risk stratification

Prognosis closely related to: • Age at diagnosis

• Extent of disease at diagnosis (presence of metastases) • Extent of residual disease after surgical resection • Histological type • Genomic and metabolic aspects increasingly important

Risk stratification

• Average risk: Children >3yrs No metastatic disease after total or near total resection • High risk: Children <3yrs (from predictions of outcome studies) Children with overt metastatic disease (CSF or imaging +ve) Residual tumor >1.5cm 2

Staging – tumor size (T)

• T1 <3cm in diameter • T2 ≥ 3cm in diameter • T3a >3cm in diameter with extension • T3b>3cm in diameter with unequivocal extension into brainstem • T4 >3cm in diameter with extension beyond cerebral aqueduct and/or down into cervical canal

Staging – metastatic disease (M)

• Mo No evidence of subarachnoid or hematogenous spread • M1 Tumor cells found in CSF • M2 Intracranial tumor beyond primary site • M3 Gross nodular seeding in spinal subarachnoid space • M4 Metastases outside the cerebrospinal axis (esp bone)

Staging – surgical residual tumor

• R0: no residual tumor • R1: residual tumor ≤ 1.5cm 2 • R2: Residual tumor > 1.5cm 2 • R3: Residual tumor infiltrating brainstem • R4: Residual tumor extending out of the posterior fossa

Risk stratification - outcome

• Non-disseminated MB patients have high likelihood of long-term survival – 80% 5-year survival • Intensified therapy increases survival in disseminated disease BUT with major quality of life issues

Radiology follow up screening

• Imaging interval dictated by trial regimes – typically 3-6 month intervals for first 5 yrs following initial therapy (for recurrence or new CSF spread)

• Imaging of brain and whole spine

• Role of long term imaging uncertain/debatable

Tumour recurrence

• Recurrence generally reflects appearance of original tumour • Some recurrences don’t enhance or may only be detected by DWI • Look for ‘hidden’ sites – anterior cranial fossa, sacral cul-de-sac

Late recurrence - nodular

April 2006 – 6 years post surgery

July 2013 – 13 years post surgery

August 2015 – 15 years post surgery

Recurrence – nodular and linear

T2

T1 post Gd

T1 post gd

T1 post Gd

Recurrence after craniospinal RT

Pre RT

Post RT

Late recurrence on DWI - linear

Presentation

10 mths post RT

15 mths post RT

ADC

T2

T1 post Gd

Late recurrence on DWI - nodular

Presentation

2 yrs later- pre SRT

1 yr post SRT

T2

T1 post Gd

DWI

ADC

Late effects

• Brain: Intellectual deficit • Ocular lens: Cataract • Retina: Radiation retinopathy • Optic nerve: Neuritis • Inner ear: Sensorineural hearing loss • Hypothalamic-pituitary axis: Endocrinopathies • Spinal cord: Chronic progressive myelitis

Late effects - radiology

• White matter damage • Radionecrosis • Tumors – meningioma • Cavernomas • Posterior fossa syndrome

White matter injury

• Post surgical damage

• Leucoencephalopathy: i. Punctate/small T2 or FLAIR hyperintensities in frontoparietal white matter ii. Confluent T2/FLAIR hyperintensities in frontoparietal WM iii. White matter cysts

Leucoencephalopathy, post surgical encephalomalacia, cavernomas

Radionecrosis

06.2015 - 5 mths post protons

12.2014 - at recurrence

11.2015 - 10 mths post protons

Late effects - meningioma

T2

T1 post Gd

T2

T1 post Gd

T1 post Gd

T1 post Gd

Posterior Fossa Syndrome

• Occurs in approx. 15-25% of midline medulloblastomas • Onset usually within 12-24 hrs of surgery • Slowly improves, but significant long term morbidity • Results from surgical damage to the efferent cerebellar pathways • Correlated with hypertrophic olivary degeneration (inferior olivary nuclei) • ? Damage to dentate nuclei/superior cerebellar peduncles • May see decreased CBF in frontal lobes

PFS imaging

1 year later

Presentation

EPENDYMOMA

Ependymoma - background

• Most common in childhood • Second most common childhood brain tumor (10%) • Occurs anywhere in neuraxis • 60% post. fossa, 30% supratentorial, 10% spinal • Account for 15% of post. fossa tumors • Most frequently diagnosed between 2 and 5 years • Slightly more common in boys

Ependymoma - background

• Arise from neuroepithelial lining of ventricles or central canal of spinal cord • Recent studies suggest radial glial stem cells as origin • Most post. fossa ependymomas arise in 4V • Most of supratentorial ependymomas are intraparenchymal, also from 3V or lateral ventricles

Ependymoma - background

• Infratentorial tumors grow exophytically from the ventricular surface of medulla: Floor of 4V (60%): extending through Magendie onto dorsal surface of cord Lateral aspect of 4V (30%): extending through Luschka into CP angle and over anterolateral aspect of pons, medulla Roof of 4V (10%) • Similar presentation to MB but usually longer duration of symptoms (6-12 months v 4 months for MB’s)

Ependymoma - Pathology

• Gd I – subependymoma and myxopapillary ependymoma • Gd II – Ependymoma • Gd III – Anaplastic ependymoma. Presence of hypercellular areas, necrosis, microvascular proliferation, high mitotic indices confer worse PFS Anaplastic histology + infratentorial location carries increased mortality risk in younger children Supratentorial location associated with higher mortality in older patients

Ependymoma - Genetics

• 2 demographic, genetic and clinical subgroups in posterior fossa ependymomas • Group A (PFA) and group B (PFB) • PFA - mainly in infants, lateral post. fossa localization: poor prognosis • PFB - mainly in older children/adults: better prognosis

Ependymoma - surgery

• GTR of tumors arising from floor or lateral aspect of 4V difficult as tumor applied to surface of brainstem and cranial nerves • Extent of resection most important prognostic factor • Intra-operative MRI optimal – enables ‘on table’ further resection • Post op imaging (24-48 hrs) to assess degree of resection • Second look surgery should be considered for residuum • Consensus MDT useful for further follow-up/surgery

Imaging

Ependymoma - CT

• Typically iso to mildly hyperdense, often heterogeneous • Approx 50% calcification (diffuse, coarse or nodular) • Cysts 20%, haemorrhage 10% • Soft, pliable tumor (“like toothpaste”) • Extends out through 4V outlets • Enhancement is variable and irregular

CT – Infratentorial Ependymoma

Pre contrast

Post contrast

Pre contrast

Post contrast

CT – supratentorial ependymoma

Ependymoma - MRI

• Heterogeneous tumor • Solid components iso- to hyperintense on T2/FLAIR • T2 hyperintense cysts frequently seen • Hypointense areas reflect calcification and/or hemorrhage • Usually iso- to hypointense on T1 • Enhancement is variable • Post. fossa tumors grow to fill 4V; may invade brainstem

Ependymoma - MRI

• 30-50% extend into the cervical spinal canal • Encasement of vessels/nerves better seen than on CT • Usually increased diffusion on DWI/patchy restricted diffusion on ADC • MRS: non-specific tumor spectra (notably high Choline)

MRI: posterior fossa - lateral

T2

T2

FLAIR

T2

T1

T1 post Gd

DWI

ADC

MRI: posterior fossa - midline

Ax T2

T1 post Gd

CISS

T1 post Gd

Ependymoma – MRS short echo

High resolution (CISS) imaging

Axial CISS

Ax T1 post Gd

MRI – supratentorial ependymoma

T2

ADC

DWI

T1

T1 post Gd

T1 post Gd

Distant spread

• Approx 15% risk of spinal spread from post. fossa tumors • Spread occurs throughout the CSF pathways • Most often nodular • Very uncommon at presentation • Infratentorial tumors have higher risk of seeding than supratentorial tumors • More common with anaplastic ependymoma • Incidence of leptomeningeal spread varies with: Tumor grade (low grade 5%, high grade 10-15%) Tumor location (5% supratentorial, 10-15% infratentorial)

Spinal metastases

T1 post Gd

T2

Tumour recurrence

• 5 yr survival: 50-64% • Recurrences typically local • Median time to recurrence 13-25 months • Distant recurrence in approx 20% • Very late recurrences (up to 20 years) not uncommon

Recurrence – Grade 2

Presentation – Aug ‘09 1 day post op

Recurrence – Mar ‘13 Post DXT + chemo – June ‘13

Recurrence – Grade 3

Jan ‘14

Jan ‘14 – 1 day post op

Jan ’15

April ‘15

May ‘15 - 1 month post re-resection

June 17

Sept 17

Cystic recurrence

T2

T2

Day 2 post op

T1 post Gd

T1 post Gd

Outcome

• 50% relapse • 25% live > 5yrs after relapse • Age impacts on treatment and outcomes • Surgery +/- chemo (commoner in early life) carries worse prognosis • Surgery +/- RT +/- Chemo (commoner in older children) carries better prognosis

Radiology screening

• Imaging interval dictated by trial regimes • Use of high resolution T2 imaging facilitates detection of residual tumour and early recurrences • Determines viability of further surgery • After end of treatment, every 3 years – late recurrences, radiation induced meningiomas

Summary

• Tumor localization and spread reflects cell origin • Increasing importance of molecular/genetic evaluation • Staging and evaluation of extent of surgery very important in disease stratification/prognosis • Tumor spread may be subtle – diffusion changes only, hidden sites (e.g. anterior skull base, spinal cul-de-sac) • MDT assessment offers optimal management • Neuraxial imaging must be the norm • Late imaging important to look for late recurrences and late effects/secondary tumors

KLINIK UND POLIKLINIK FÜR STRAHLENTHERAPIE UNIVERSITÄTSKLINIKUM LEIPZIG

Part I :

General aspects

Clinical features, histology, surgery, staging, prognostic factors, outcome

R. Kortmann / B. Timmermann

Ependymoma

Distribution

Intracranial (60%) Supratentorial : 30% Infratentorial : 70%

Metastases Extramedullary < 5 – 10%

4-16 years (ca) Supratent : 35% Infratent : 50%

Spinal (30%) Intramedullary (thoracal) (10%) Extramedullary (lumbar) (20%) 4-16 years (ca) Intramed : 10% Conus 5%

Schiffer et al., 1991

Ependymoma

Ependymoma Spinal metastases

Ependymoma

Intracranial and spinal ependymoma SEER Analysis / n=354 patients (children only)

5 year overall survival

Supratent : 106 pat. Infratent.: 193 pat.

57.8% 54.4% 86.6%

Spinal : 55 pat.

McGuire et al., 2009

Ependymoma

Localisation of tumour

Infratentorial

Supratentorial

Supratentorial (interventricular often in adults)

Ependymoma

Therapeutic strategy

Surgery

extent of resection

Postoperative RT Chemotherapy

local RT / entire PF /CSA

protocol

Supratentorial

Infratentorial

Intracranial ependymoma

Postoperativ radiotherapy Criteria for therapeutic decision

Localisation

supra- / infratentorial

WHO Grade

WHO Gr. II WHO Gr. III (anapl.)

Extent of resection

complete / incomplete

Metastases

no / yes

Age

< 3-5 years > 3-5 years

Intracranial ependymoma

Clinical features

Staging

Intracranial ependymoma

Age and sex distribution

Author (series)

Sex

Age

McGuire et al., 2009 (SEER) (55 spinal tumours included / 8.7%)

Male : 370 (58.3%) Female : 265 (41.7%)

0-4 years : 329 (51.8%) 4-18 years : 306 (48.2%) < 3 years : 78 (51%) >/= 3 years : 75 (49.0%)

Merchant et al., 2009 (St. Jude)

Male : 95 (62.1%) Female 58 (37.9%)

Intracranial ependymoma

Clinical features

Depending on location

1. Posterior fossa tumours : - raised intracranial pressure - visual disturbances - ataxia and hemiparesis

- dizziness - neck pain - cranial nerve palsies.

2. Supratentorial tumours - headache, seizures - focal neurologic deficits depending on region involved

Intracranial ependymoma

Staging

Staging before radiotherapy (before surgery)

- Pre- / postoperative MR (brain) - MR of spinal canal (before surgery and before lumbar puncture)

Intracranial ependymoma

Prognostic factors

Intracranial ependymoma

Prognostic factors Tumour site (infratentorial / supratentorial) progression – free survival at 5 years

Author

Pat.

Tumour site Survival

p-value

Schild et al., 1998

45

Infratent. 68% Supratent. 62% Infratent. 53.1% Supratent. 72.4% Infratent. 65.8% Supratent. 82.9% Infratent. 42.5% Supratent. 50.9% Infratent. 65.2% Supratent. 31.3%

n. s.

Timmermann et al. 2000*

29

n. s.

26

Merchant 2008 et al.**

122

0.16 / n. s.

31

Jaing et al.,2004

28

n. s.

15

Mansur et al.,2005

48

n. s.

12

* 3 year event-free survival, ** 7 year event-free survival

Intracranial ependymoma

Prognostic factors / age cut – off : 16 years

supratentorial : 17 pat. , infratentorial : 22 pat. , spinal : 15 pat.

1

actuarial survival 5 years : 72% 10 years : 58%

0.8

>16 years n = 33

0.6

0.4

probability

<= 16 years, n = 21

actuarial survival 5 years : 40% 10 years : 40%

0.2

p=0.03

Stüben et al., 1997

0

0

5

10

15

20

25

years

Intracranial ependymoma

Prognostic factors / age cut : off : 10 years CCSG random. study / survival by age

1

actuarial survival 10 years : 57%

0.8

10 pat.

0.6

10+ years

0.4

probability

26 pat.

1-9 years

0.2

actuarial survival 10 years : 31%

p = 0.11

Evans et al., 1996

0

0

3

6

9

12

15

years

Intracranial ependymoma

Progression-free survival / by age Prognostic factors / age cut – off : 3 years

1

0.8

0.6

> / = 3 years / n = 31

0.4

probability

0.2

P = 0.019

< 3 years / n = 18

0

0

60

120

180

240

Shu et al., 2007

months

Intracranial ependymoma

Prognostic factors / age cut – off : 3 years

Overall survival / by age

1

0.8

> / = 3 years / n = 31

0.6

0.4

probability

0.2

P = 0.006

< 3 years / n = 18

0

0

60

120

180

240

Shu et al., 2007

months

Intracranial ependymoma

Histology Prognostic factors

Controversial results

➢ Classification

➢ Institutional policies

Need for standards

Intracranial ependymoma

Prognostic factors / grading disease-free survival

1

0.8

Grade II / n = 40 DFS 5 years : 61.5% DFS 10 years : 54.9%

0.6

0.4

Grade III / n = 20 DFS 5 years : 51.6% DFS 10 years : 36.9%

Probability

0.2

p = 0.62

Mansur et al., 2005

0

0 50 100 150 200

250 300 350 400 450

months

Intracranial ependymoma

Prognostic factors

AEIOP OS/PFS according to histological grading

1

74%

0.8

66%

0.6

p < 0.0001

0.4

25%

OS classic (n=43) PFS classic OS anaplastic (n=20) PFS anaplastic

probability

0.2

8%

0

0 12 24 36 48 60 72 84

months

Massimino et al., 2004

Intracranial ependymoma

Histology / discrepancies in diagnosis Prognostic factors

SFOP 73 Babies 27 months (5-62)

UKCCSG 73 Babies 19 months (4-38)

Classic WHO grade II Anaplastic WHO grade III

74%

17%

83%

13%

Awaiting review

-

12%

Intracranial ependymoma

Histological assesmant of grading / discrepancies between pathologists / european panel

Overall survival with respect to investigator

Pathologist A

Pathologist B

1.0

1.0

0.8

0.8

Pathologist A

0.6

0.6

0.4

0.4

Grade II Grade III

Grade II Grade III

0.2

0.2

0.0

0.0

0

2

4

6

8

10

12

0

2

4

6

8

10

12

Time (Years)

Time (Years)

Intracranial ependymoma

Impact of postop. RT on outcome Overall survival / infratentorial tumours / SEER data bank

RT : 116 pat. 57.1% No RT : 68 pat. 48.2% P = 0.018

McGuire al., 2009

Intracranial ependymoma

Relapse-free survival / extent of disease Anapl. ependymoma / HIT 88/89/91

1

0.9

Localis. (n=50) Metast. (n=5)

0.8

0.7

0.6

65.8%

0.5

0.4

0.3

0.2 probability

p = 0.0001

0.1

Timmermann et al., 2000

0

0

20

40

60

months

Intracranial ependymoma

Impact of postop. RT on outcome (54 Gy) Actuarial local control rates

1

GTR / RT : n = 13 PFS 10 years : 100%

0.8

GTR alone : n = 19 PFS 10 years : 50.0%

0.6

0.4

Probability

GTR alone versus GTR + RT : GTR + RT versus STR + RT :

STR / RT : n = 12 PFS 10 years : 36.0%

p = 0.018

0.2

p= 0.003

Rogers et al., 2005

0

0

24

48

72

96

120

months

Intracranial ependymoma

Impact of postop. RT on outcome (54 Gy) overall survival

1

GTR / RT : n = 13 OS 10 years : 83%

0.8

GTR alone : n = 19 OS 10 years : 67%

0.6

0.4

Probability

GTR alone versus GTR + RT : GTR + RT versus STR + RT :

STR / RT : n = 12 OS 10 years : 43%

p = 0.507

0.2

p= 0.088

Rogers et al., 2005

0

0

24

48

72

96

120

months

Intracranial ependymoma

Postoperative radiotherapy / 5 year survival

Author

Pat.

Surg. only Surg. + RT

Mork

12 16 7 10 16 74 15 65 31 12

17% - 18% -

- 40% - 68%

Ferrante

Perilongo*

20.4% -

- 38.2%

Rousseau*

0% -

- 45%

Jaing

48.6% -

- 57.9%

* Event – free survival

Intracranial ependymoma

Anapl. ependymoma / HIT 88/89/91

Relapse – free survival / residual tumour

1

Compl. res. (n=28) Incompl. res. (n=27)

0.9

83.3%

0.8

0.7

62.4%

0.6

0.5

0.4

0.3

38.5%

30.8%

p = 0.0043

0.2

probability

0.1

Timmermann et al., 2000 Timmermann et al., 2004

0

0

20

40

60

months

Intracranial ependymoma

Extent of resection / St. Jude series / event-free survival

(GTR=117) 5 YR EFS 78% + 4% (NTR=12) 5 YR EFS 65% + 16% (STR=11) 5 YR EFS 65% + 16%

Log-Rank p=0.017

Thomas Merchant, Paris, 2007

Intracranial ependymoma

HIT 2000 hfx / Extent of res. / EFS / OS / Gr II/III - R0 vs. R+

EFS

OS

survival function

survival function

R+ n =14 Pat. 3 y OS 85.7% 5 y OS 85.7% R0 n =54 Pat. 3 y OS 92.6% 5 y OS 90.7 %

R0 n =54 Pat. 3 y EFS 87% 5 y EFS 77.9 %

Cumulative survival

Cumulative survival

R+ n = 14 Pat. 3 y EFS 50%; 5 y EFS 25 %

p = 0.550

p = 0.001

years

years

Kortmann et al., ISPNO 2010

Intracranial ependymoma

Prognostic factors extent of resection and Grading AEIOP series (Massimino et al., 2004) / overall survival

CR / GD II CR / GD III STR / GD II STR GD III

P = < 0.0001

Ellison et al., 2011

Intracranial ependymoma

Adverse prognostic factors 3 yr event-free survival / european data

Subtotal Resection – 53% vs. 83%

p=0.029

Anaplastic Tumor Grade – 50% vs. 94%

p<0.0001

Pre-irradiation Chemotherapy – 54% vs. 86%

p=0.0017

Multivariate Analysis –

Extent of Resection

p=0.018 p=0.0003 p=0.0106

– –

Tumor Grade

Pre-irradiation Chemoth.

Molecular genetic profiles and prognostic implications Location posterior fossa Intracranial ependymoma

Ramaswamy et al., 2016

Molecular classification of ependymal tumors across all CNS compartments, histopathological grades, and age groups Intracranial ependymoma

Pajtler et al., 2015

Molecular classification of ependymal tumors across all CNS compartments, histopathological grades, and age groups Intracranial ependymoma

Posterior fossa

supratentorial

EPN-PFA EPN-PFB

EPN-YAP

EPN-RELA

5-year PFS 5-year OS Nb of pat.

33%

73%

66%

29% 75%

68% 100%

100%

240

51

13

88

Pajtler et al., 2015

Intracranial ependymoma

Molecular genetic profiles and prognostic implications PF location / 1q25 gain / Tenascin C / OS

Andreiuolo et al., SIOP Ependymoma Working Group, 2017

Intracranial ependymoma

Molecular genetic profiles and prognostic implications PF location / 1q25 gain / Tenascin C / OS

Andreiuolo et al., SIOP Ependymoma Working Group, 2017

Intracranial ependymoma

Molecular genetic profiles and prognostic implications Supratentorial location / 1q25 gain / Tenascin C / OS

Andreiuolo et al., SIOP Ependymoma Working Group, 2017

Intracranial ependymoma Molecular genetic profiles and prognostic implications Supratentorial location / 1q25 gain / Tenascin C / OS

In supratentorial tumours ca 75% RELA fusion

Andreiuolo et al., SIOP Ependymoma Working Group, 2017

Intracranial ependymoma

Molecular genetic profiles and prognostic implications

Pajtler et al., 2017

Intracranial ependymoma

Management in the infant

Intracranial ependymoma

Anapl. ependymoma / HIT SKK

Role of RT in infants (< 3 years) / PFS - OS

1

0.8

Overall survival

0.6

3-yrs.-pfs = 14.9 % 3-yrs.-os = 17.2 %

0.4

Probability

0.2

Progression free survival

Timmermann et al., 2004

0

0

20

40

Survival (months)

Intracranial ependymoma

Anapl. ependymoma / HIT SKK

Role of RT in infants (< 3 years) / RT yes - no

1

0.9

0.8

Irradiated ( n=14 ) Not irradiat. ( n=15 )

0.7

0.6

0.5

0.4

Probability

0.3

0.2

Timmermann et al., 2004

p = 0.02

0.1

0

0

20

40

Overall survival (months)

Intracranial ependymoma

Anapl. ependymoma / HIT SKK

Role of RT in infants (< 3 years) / up front RT yes - no

1

0.9

Preventive RX (n= 10) Salvage RX (n= 4)

0.8

0.7

0.6

0.5

0.4

Probability

0.3

0.2

p = 0.13

0.1

Timmermann et al., 2004

0

0

20

40

Overall survival (months)

Intracranial ependymoma

Impact of timing of RT on survival

Overall survival / by RT intent

1

0.8

Initial RT / n = 34

0.6

0.4

probability

Delayed RT / n = 13

0.2

P = 0.499

0

0

60

120

180

240

months

Shu et al., 2007

Intracranial ependymoma

Anapl. ependymoma / HIT SKK

Timmermann et al., 2004

Intracranial ependymoma

HIT 2000 / age : 4 – 21 years / M0 Ependymoma ° II/ ° III

RT tumour site only 68 Gy, 2 x 1.Gy / day (boost 72 Gy in persistent residual disease) weekly VCR in Gr. III

In grade III Chx HIT SKK 2000* 5x alternating CP/VCR

Surgery

CARBO/VP16 *no i.th. MTX

M 1 – 3 : RT CSA : 40 Gy hfx (2 x 1 Gy / day), boost tumour site 68 Gy, 72 Gy in persistent Tu., 50 Gy to spinal deposits + Chx. HIT SKK

HIT 2000 / age : < 4 years / M0

RT tumour site only 54 Gy, 5 x 1.8 Gy / week (stereotactic boost in pers.disease)

Regardless of grade Chx HIT SKK 2000*

Surgery

M 1 – 3 : RT CSA : 24 Gy, boost to tumour 54 gy + deposits 44.8Gy

Intracranial ependymoma

EFS - Ependymoma ° II+ ° III <4 years / R 0 versus R +

R0: < 4 years 5 y EFS : 84.3 % n = 61, 9 relapses

R1 : < 4 years 5 y EFS : 63.5 % n = 25, 9 relapses

survival

p = 0.016

years

Intracranial ependymoma

Ependymoma – international data / „infants“

Author, year

Pat.

RT

Survival

Grill et al., 2001

73 Chx only / RT at PD

4 y PFS : 26%/ 4 y. OS : 59% (23% without RT)

Grundy et al., 2007

89 Chx. only RT at PD

M0 EFS OS 3 y 47.6 79.3% 5 y 41.8% 63.4%

Fouladi et al., 2009

21 Planned RT

5 y PFS : 33% 5 y. OS : 62%

< 18 mon : 48Gy 18-30 mon. :51 Gy > 30 mon. : 54 Gy

Massimino et al.,

41 Chx. only / RT at PD

5 y PFS : 26%/ 5 y. OS : 37%

Merchant et al., 2009

78 RT only

5 y EFS : 68.6%/ OS : 80.4%

Grundy et al., 2010

11 Chx. only / RT at PD

1 y PFS : 1/11/ 1 y. OS : 9.1%

Timmermann et al., 2004 HIT 88-89/91

34 RT : 21

3 year PFS : 23.3% 3 year OS RT : 66.7%/ no RT : 38.5%

No RT : 13

HIT 2000

51 R0 : 61 R+ : 25

5 year PFS R0 : 84.3 / R+ : 63.5%

Intracranial ependymoma

Role of chemotherapy

Intracranial ependymoma

CCSG random. study / survival by arm (+/- chx.

1

0.8

actuarial survival 10 years : 40%

0.6

RT + Chx. (n = 22)

0.4

probability

RT only (n = 14)

actuarial survival 10 years : 35%

0.2

p = n.s. (0.93)

0

3

0

6

9

12

15

years

Evans et al., 1996

Intracranial ependymoma

HIT 91 random. study / Sandwich vs. maintenance Event-free survival / R0 / n = 26

1

Maintenance – chx. (8 x cispl, VCR, CCNU)

n = 14, 79% +/- 11%

0.8

n = 14, 65% +/- 14%

0.6

Sandwich – chx.

0.4 probability

Immediate pop. RT after complete resection !

0.2

p = n.s.

0

years

3

0

6

9

12

HIT – study group, 2007, unpublished

Intracranial ependymoma

HIT 91 random. study / Sandwich vs. maintenance

Event-free survival / R1 / n = 18

1

0.8

Sandwich – chx.

n = 13, 46% +/- 14%

0.6

0.4 probability

Maintenance – chx. (8 x cispl, VCR, CCNU)

0.2

n = 5, 0%

Up-front chx. after incomplete resection ?

p = 0.002

0

years

3

0

6

9

12

HIT – study group, 2007, unpublished

Intracranial ependymoma

CCG 9942 phase II pre-Irradiation chx in incompletely resected ependymoma (=/> 3 years) n= 41, n=43 RT alone after compl. Resection Chx : VCR, Cisplat, VP 16, Cycloph. (RT local 54-59.4Gy

Response to chx. : 35 pat. : CR 14 (40%), PR 6 (17%) MR/SD 10 (29%, PD 5 (14%)

Garvin et al., 2012

Intracranial ependymoma

Summary

Prognostic factors

➢

extent of resection,

➢

age,

➢

grading (consensus !) future role ?

➢

molecular genetic markers / stratification

Dose – response relationship

➢

> 54 Gy

➢

role of hfx unclear (studies)

➢

duration of overall treatment time

RT of tumour site (CSA in M+ disease)

➢

3 – D conformal technique

➢

radiosurgery (hypofractionated ?)

RT in children < 3-5 years

➢

immediate RT (?)

Role of Chx. unclear (studies)