Screening for oral cancer

Question 1

Define ‘screening’ (2)

Answer 1

‘the application of a test or tests to people who are apparently free from the disease in question in order to sort out those who probably have the disease from those who probably do not.’

‘A screening test is not intended to be diagnostic’

-interrupt natural history of disease at asymptomatic stage when it is treatable and progression can be halted

Question 2

Screening programmes (3)

Answer 2

Organised screening for disease, including provision for recall, referral, specialist treatment and evaluation
Screening is a continuing process
People are screened at regular intervals

Question 3

Types of screening (3)

Answer 3

Mass (population) screening
-large scale screening of population groups
-usually by individual
Selective screening
-targeted screening of high-risk groups
Opportunistic screening
-examining individuals when they attend for some other, often unrelated, purpose

Question 4

The 10 principles of screening (10)

Wilson and Junger 1968

Answer 4

The condition is an important health problem
Its natural history is well understood
It is recognisable at an early stage
Treatment is better at an early stage
A suitable test exists
An acceptable test exists
Adequate facilities exist to cope with abnormalities detected
Screening is done at repeated intervals when the onset is insidious
The chance of harm is less than the chance of benefit
The cost is balanced against benefit

Question 5

Potential advantages of cancer screening (7)

Answer 5

Reduced mortality
Reduced morbidity
Reduced incidence of invasive cancers
Improved prognosis for individual patients
Identification of high-risk groups and opportunities for primary intervention
Reassurance for those screened negative
Cost savings

Question 6

Potential disadvantages of cancer screening (6)

Answer 6

Detection of cases already incurable may increase morbidity for some patients
Unnecessary treatment for lesions which may not have progressed
Psychological trauma for those with a false positive screen
False reassurance for those with a false negative screen
Reinforcement of bad habits among those screened negative
Costs

Question 7

Screening programmes that work (3)

Answer 7

Cervical cancer
-25yrs onwards every 3 years
Breast cancer
-50-70yrs every 3 years
Colon cancer

Question 8

Cervical cancer screening (5)

Answer 8

Women aged 25 - 64
Cervical smear detects abnormal cells and Human Papilloma Virus
If positive: patient referred for specialist exam and biopsy
Re-screened every 3 years (up to 50) and then 5 years
Cost - £175m per year

Question 9

Breast cancer screening (7)

Answer 9

Started in 1988
Women 50-70 years
Mammogram to detect subclinical abnormalities
If positive: patient referred for specialist exam and biopsy
Re-screened every 3 years
Cost - £100m per year
2 - 2.5 lives are saved for every overdiagnosed case (Duffy et al 2000)

Question 10

Bowel cancer screening (5)

Answer 10

Males and females 60 – 75 yrs, every 2 years (will soon be extended to 50-75 year olds)
Home test for faecal occult blood to detect subclinical abnormalities
If positive: patient referred for specialist exam and biopsy
Started in 2006, fully operational from 2010
Cost - £85m per year

Question 11

Screening programmes that do not work - prostate (3)

Answer 11

Disease detected too early, when progression may not affect health within lifetime
High false positives- over treatment
Test is offered to worried men

Question 12

Screening programmes that do not work - lung (2)

Answer 12

Disease detected too late, when patients will progress irrespective of treatment
Good example of lead – time bias

Question 13

Lead-time bias (3)

Answer 13

Screening should increase survival time
Positive screen comes before symptoms
You know about it for longer in long cancer, but does not mean that you will live any longer
Cervical cancer - you have a big true increase in survival

Question 14

Criteria for screening - oral cancer (5)

Answer 14

Disease must be common and serious
-relatively common
-incidence is increasing
-people typically present late
Disease must have a known natural history
- keratosis –> dysplasia –> carcinoma
-leukoplakia and erythroplakia easy to spot
-but we do not accurately know which lesions will progress
A good screening test must be available
-yes but more research is needed
-current research directed at using brush biopsy cytology to identify screen-detected lesions that are most likely to be dysplastic
Effective treatment must be available
-early detection improves survival by a lot
It must be cost-effective
-no

Question 15

Leukoplakia and malignancy (4)

Answer 15

Overall about 5% become malignant within 5 years
About 1.5%/ year transformation
95% do not progress - but which ones
Still no reliable way to predict which individuals or lesions will develop carcinoma

Question 16

What would the screening test for oral cancer be? (4)

Answer 16

Systematic visual examination of the oral mucosa
‘Positive’ screen
-a white patch
-a red patch
-an ulcer of longer than 2 weeks duration

Question 17

Does oral cancer screening work? (4)

Answer 17

Need a randomised controlled trial with mortality as the end point
Kerala study conclusions at 9 years:
-oral visual screening can reduce mortality in high-risk individuals
-potential to prevent > 37000 oral cancer deaths
–>high-risk individuals: smokes >20 per day and drinks over limit

Question 18

Studies done in UK

Answer 18

Screening questionnaire

• 2336 participants
• full habits and lifestyle questionnaire
• examined by a general dentist (screener)
• second blind examination and diagnosis by a specialist in Oral Medicine – gold standard

Only 3 had the cancer, although 71 (3%) tested positive
-quite low sensitivity

Question 19

Oral cancer screening meta-analysis of all screening studies - stats (5)

Answer 19

85% of lesions correctly identified
15% false negatives – missed cancers!
3% false positives
70% likelihood of being right
1% likelihood of being wrong

Question 20

Treatment for oral cancer (3)

Answer 20

Prognosis is generally poor, with less than 54% overall 5 year survival
Associated with radical surgical treatment unless detected and treated early
If detected early prognosis and outcome are excellent with 90% five-year survival for stage 1 lesions.

Question 21

Cost-effectiveness (3)

Answer 21

In the Kerala study they also demonstrated cost-effectiveness
It is not cost effective to screen whole population in the UK
In a UK population, simulation modelling suggests that opportunistic screening in primary care may be cost-effective