De asymptomatische screenings - PowerPoint Presentation

Slide1

De asymptomatische screenings patiënte. Kan het zonder echografie?

Dirk Timmerman

VVOG Genk, 11

oktober

2013

CME:

This

presentation

is free

from

any

commercial bias and /

or

promotional

intent

Slide2

Endometrial CarcinomaPremenopausal:1) Endometrial thickness varies in relation to menstrual cycle (Day 4-8!)2) Benign endometrial pathology will be more common3) Endometrial cancer would be expected to be lower

Pos. predictive value will be lower

Cutt

-off 5 mm

Postmenopausal

Premenopausal

Sensitivity

95 %

80 %

Specificity

55 %

34 %

Slide3

Prognosis of Endometrial Cancer Screen detected vs Symptomatic Lesions: No prognostic

advantage

Gerber et al. EJC 2001

16 screen detected vs

190 symptomatic endo ca 5-year DFS OS

No Bleeding: 77% 86%

< 8 Weeks Bleeding 83% 98%

8-16 Weeks Bleeding 74% 90%

>16 Weeks Bleeding 62% 69%

Slide4

Screening (WHO criteria)High prevalence

and serious 

Spectrum of symptoms

is

known Screening should be

easy



few FP

Few FN Methods for further confirmation exist Diagnosis will lead to improved therapy Cost

Slide5

ConclusionNot any trial of screening for endometrial cancer has shown a positive effect on mortality

Many false positive results lead to a large number of interventions and create much morbidityRoutine screening of all (postmenopausal) women would be very costful

Patients at higher risk avoid screening

Slide6

Ovarian cysts and tumors

Slide7

Pre- vs. postmenopausal womenInvasive

ovarian cancer

Invasive

ovarian

cancer

Slide8

Ovarian cancerLeading cause of death from gynecological cancers> 50% mortality rate“Silent killer”: early disease is asymptomatic or non-specific nature of symptoms and signs10% diagnosed at localized stage I – 90% 5Y survival rateMost women are diagnosed with advanced disease1/70 life-time risk60.000 new patients per year in Europe; 22,220 new cases in US

Slide9

Ovarian Cancer

ScreeningDiagnosis

Risk assessment

Monitoring

FDA approvedOVA1CA125

HE4

On

market

OvPlex

Ultrasound

ROMA

Clinical practice(CA 125Ultrasound)CA 125UltrasoundRMI / CT / MRIIOTA simple US rulesCA 125(ultrasound)CT/ MRIIn develop-mentROCA (MMS)Ultrasound ++OvaSure

OvaDxOvaCheck

IOTA multi-class

models (US)

Slide10

Ovarian Cancer

ScreeningDiagnosis

Risk assessment

Monitoring

FDA approvedOVA1CA125

HE4

On

market

OvPlex

Ultrasound

ROMA

Clinical practice(CA 125Ultrasound)CA 125UltrasoundRMI / CT / MRIIOTA simple US rulesCA 125(ultrasound)CT/ MRIIn develop-mentROCA (MMS)

Ultrasound ++OvaSureOvaDx

OvaCheck

IOTA multi-class

models (US)

Slide11

Rationale for screening for ovarian cancerIII

III

IV

STAGE

%% EACH STAGE

% 5 YR SURVIVAL

Slide12

BIOPHYSICAL MARKERSBIOCHEMICAL MARKERSScreening tests

TUMOUR MARKERSCA 125HE4

OVA1

Imaging

TV ultrasoundColor Doppler

3D

Slide13

Screening: 1 ovarian cancer/1000

Normal

/

benign

999

Slide14

Screening: 1 ovarian cancer/1000

Normal

/

benign

999

Ovarian

cancer

1

Slide15

Screening. Test performance?

Sensitivity

100%

Specificity

95%

50

operations

/1cancer

Slide16

Screening. Test performance?

Sensitivity

100%

Specificity

99%

10

operations

/1cancer

Slide17

IOTA: preoperative assessment of postmenopausal patientsInvasive 367

Benign

515

Borderline

66

Metastatic

53

Slide18

IOTA: Benign vs not benign Invasive 367

Benign

515

Borderline

66

Metastatic

53

Laparoscopy

Oncology

treatment

Slide19

Screening: detection of invasive cancerInvasive 367

Benign

515

Borderline

66

Metastatic

53

False

positives

True

positives

Slide20

Screening vs. Preoperative diagnosisUKCTOCSIOTA 1, 1b, 2, 3

Slide21

Screening vs. Preoperative diagnosisUKCTOCS

IOTA 1, 1b, 2, 3

Slide22

Screening studies in general populationCA 125 +/- USSEinhorn et alJacobs et al

Grover et alAdonakis et al

49,590 women

28 ovarian cancers detected

USS +/- CDIVan

Nagell

/

DePriest

et al

Sato et al

Hayashi et al

Campbell et al

Schingalia et alParkes et alVuento et al Goswamy et al110,322 women 47 ovarian cancers detected

Slide23

Ultrasound criteria in screening studiesOvarian volumePLCO : CA-125 ≥35 U/mL: abnormal(1) ovarian volume greater than 10 cm3, (2) cyst volume greater than 10 cm3, (3) any solid area or papillary projection extending into the cavity of a cystic ovarian tumor of any size; (4) any mixed (solid and cystic) component within a cystic ovarian tumorUKCTOCS: abnormal scan: one or both ovaries had complex morphology or simple cysts greater than 60 cm³, or ascites

Slide24

Is surgical resection of benign ovarian tumors useful to prevent ovarian cancer? 9 women treated with ovarian cancer

202 bilateral oophorectomies for benign lesions Follow-up for 15.5 years

Ovarian cancer mortality

Predicted Observed 25 22

(Crayford et al 2000)

(Courtesy Prof Lil Valentin)

Slide25

Bart’s / Royal London Study

22,000postmenopausal

womenprevalencescreen

Study

Group10,958Annualscreening 3yControl

Group

10,977

No

screening

Follow up - Mean 8y

Questionnaire & Cancer Registry

Slide26

Bart’s / Royal London trial

9

6

8

4

7

2

6

0

4

8

3

6

2

4

1

2

0

1

.

0

.

8

.

6

.

4

.

2

0

.

0

Cumulative Survival

Months post randomisation

p=0.011

Control

Group

(n=20)

Screened

Group

(n=16)

Median Survival 73mths vs 42mths

(Jacobs et al, Lancet 1999;

353:1207

)

Slide27

Is ovarian cancer a suitable disease for screening?

NORMAL

OVARY

PRE-

MALIGNANT CHANGES

PRE-

CLINICAL DISEASE

Does early detection improves prognosis?

Lead time

bias

Length time bias

No

recognised premalignant stageDuration unknown

(Courtesy U

Menon)

CLINICAL DISEASE

Slide28

Ovarian cancerLead time bias Length time bias

…earlier detection …slow growing cancers with good not altering the time of death prognosis are detected at screening

Screen

Death

Stage

I

II

III

IV

45

52

65

61

Symptoms

Screen

Screen

Death

Stage

IV

III

II

I

45

(Courtesy Prof Lil Valentin)

Slide29

TrialMethodN of screensAgeCompletedERTOCSUS

2 or 350-64StoppedPLCO

Phys exam, US + CA125

460-74

2011UKCTOCSUS or CA125650-742015

Randomised trials on ovarian cancer screening

Slide30

Slide31

Cumulative deaths

Slide32

Conclusions from PLCO trialAmong women in the general US population, simultaneous screening with CA-125 and transvaginal ultrasound compared with usual care did not reduce ovarian cancer mortality. Diagnostic evaluation following a false-positive screening test result was associated with complications.

Slide33

Retrospectively evaluated serial preclinical serum CA125 values measured annually in 44 incident ovarian cancer cases identified from participants in the PLCO trialLongitudinal algorithm identified ovarian cancer 10 months earlier and at lower CA125 concentrations (20 U/mL) than a single-threshold screening algorithm (≥35 U/mL)

Slide34

Slide35

Screening for ovarian cancerVan Nagell2007Kobayashi2008PLCO2005

UKCTOCSUS groupUKCTOCSMMS groupStudy design

Single armRCTRCT

RCTRCT

StrategyUltrasoundPhys. exam, US + CA125US + CA125USCA125 (ROCA)N of

women

25327

41688

28816

48227

50078

N of

screens4.85.41 (first)1 (first)1 (first)N mal (border)39 (10)27 (-)27 (9)45 (20)42 (8)N surgeries36490357084597Sensitivity76%(77%)-75%89%Specificity98.7-98.4%98.3%

99.9%N oper/

cancer933

21353

% stage I/II ca82%-

22%50%47%

(Lancet

Oncol

2009)

Slide36

Impact on mortality is not known yetTotal cost cannot be calculated (MMS: patented)No clear guidelines given for USUS is very sensitive for benign tumours (no benefit in removing benign tumours)Bias in favour of MMS? Changing of cut offs in MMS group DURING the study

Slide37

Screening (WHO criteria)High prevalence

and serious 

Spectrum of symptoms

is

known Screening should be

easy



few FP



Few FN Methods for further confirmation exist Diagnosis will lead to improved therapy Cost

Slide38

Screening for ovarian cancer:conclusionsAt present screening of the general population cannot be recommendedNeed to prove decrease of mortality from ovarian cancer in the screened groupFurther studies are needed to improve screening algorithms: better diagnostic tests?

Slide39

Ovarian cancer: the silent killer?JournalYearStudy

TypeN casesN

controls

Symptoms ov

caSymptoms controlGynecol Oncol

1999

Retrospect

Early

stI-II

50

ov

ca22BOT82% (3.4mths)Pain, bloating, vaginal bleedingBOT: 68% (8mths)Pain, bloating,vaginal bleedingJAMA2004Prospect44Ov ca1709controlsIncreased abdominal size (OR 7.4), bloating (OR 3.6), urinary urgency(OR 2.5), pelvic pain (OR 2.2)Healthy: Back pain, fatigue, bloating , abdom pain, constipGynecolOncol2004Prospect665Ov ca

146BOT96%

of stage III-IV93% of stage I-II

Abdominal pain (44%), bloating (39%), mass (12%)

84% of BOTLess general malaise

Cancer 2005

Retrospect19856024 br

ca10941 nl30% abdom

pain (OR 6),

16%

abdom

swelling

(OR 31),

8%

gastrointest

sympt

(OR 2),

5%

pelvic

pain (OR 4)

Symptoms

more

than

6

months

before

diagnosis

J

Womens

Health

2007

Retrospect

920

2760

36%

abdominal

symptoms

10%

genital

symptoms

13% urethra/

urinary

symptoms

7%

3%

6%

Eur

J

Gynaecol

Oncol

2007

Prospect

244

151

benign

92%

:

weight

loss/

gain

,

general

malaise,

chest

pain,

abd

swelling

,

bowel symptoms

GynecolOncol2008

Prospect4459 benign21 normal

Persistent abdom distention (OR 5)PMB (OR9), appetite loss (OR 3)

Slide40

Symptoms of ovarian cancerAbdominal pain (persistent) Abdominal swellingBloating, loss of appetiteWeight reduction or increaseMalaiseUrinary symptoms (urgency, urethra)Pelvic painVaginal bleedingThoracic painNot: back pain, constipation, fatigue…

Slide41

Preoperative assessment of ovarian tumors

Slide42

Slide43

LR2 prediction model

Blood flow in a papillary structure

Max size of solid component

Irregular internal cyst walls

Ascites

Acoustic shadows

z

= −5.3718 + 0.0354(1) + 1.6159(2) + 1.1768(3) + 0.0697(4) + 0.9586(5) −2.9486 (6)

AGE

Cut off ≥ 10%

Not: color score, max diameter of lesion, purely solid lesion...

Slide44

Role of CA 125 ?

Slide45

Slide46

Histological diagnosisCA-125 related to tumour stageAbscess, fibromas,

endometriomas have higher CA-125; comparable to borderline tumours

Slide47

Conclusions of serum CA 125 for preoperative classificationCA 125 has no added value to predict malignancy in adnexal massesIn multivariate modelsIn comparison to subjective impressionAs a second stage test (subjective assessment)In specific cases (borderline, stage I, difficult)

Slide48

Prospective study on 389 patients with a pelvic mass, who were scheduled to have surgery. The performance of each of HE4 and CA125 as well as that of ROMA, was analysed.

Slide49

Slide50

Slide51

these 10 rules were applicable to 1501/1938 (77.0%) tumorssensitivity 92% (pattern recognition 91%)specificity 96% (pattern recognition 96%)LR+ 21.29LR- 0.08

Slide52

Green-top Guideline November 2011

Slide53

User-friendly IOTA apps

Slide54

Multiclass model as reliable as IOTA LR

(Van

Calster B et al,

BMC Medical Res Methodol, 2010

)DataAll IOTA data

3511 patients

21 centers

Slide55

Conclusies screening naar

ovariumcarcinoom

Nut van screening

is nog

niet bewezenHet is wel zeer

belangrijk

om

klachten

ernstig

te

nemen

en gyn echo uit te voerenNieuwe studies verdienen betere technieken:simple ultrasound-based rules subjectieve beoordelingMulti-class modellenDank u!