Towards Earlier Cancer Diagnosis

Download Report

Transcript Towards Earlier Cancer Diagnosis

Towards Earlier Cancer
Diagnosis
Hamish Whitaker GP Tutor
Guildford
Context
• 25% deaths in developed world from
cancer
• Most present to primary care with
symptoms, even if a screening test exists
• Early symptoms shared with benign
disease
• Average full time UK GP will have a new
cancer diagnosed in one of their patients
each month
Context
• UK cancer mortality record poor compared
to other countries
• Current estimate 7500-10000 lives lost
annually due to late diagnosis
• Two week rule and easier access to
investigations has made little difference
Why the delay?
• Patient delay – eg only 41% patients report
rectal bleeding to GP
• Positive predictive value of colorectal cancer
with bleeding is 0.1%- but 2-3% by the time
presented to GP (and 5-7% by time get to 2ary
care)
• Do UK GPs have too high a threshold for
selecting patients for referral?
• Pressure from PCT and 2ary care not to refer,
criticism of inappropriate use of guidelines
Breast cancer
• 75% breast cancers identified following
presentation of symptoms
• Screening much less impact
• Little evidence to guide referral exists
• Breast pain and lumps commonest
symptoms
• 8% women with breast lump reported to
primary care are cancer- but depends on
age (0.5% if < 25 yrs)
Breast cancer
• Breast pain cancer risk 0.9% overall
• Younger women higher risk of
inflammatory breast cancer
• Present with swelling redness and
warmth, may appear rapidly and therefore
be mistaken for infection
• Poor prognosis
Lung cancer
• Prognosis generally poor – does early
diagnosis make a difference?
• Usually spread by the time symptoms
presented
• Only 20% UK patients eligible for surgical
resection
• Screening little impact – current study on
annual cytology and CT in COPD
smokers, and spiral CT
Lung cancer
• Italian spiral CT trial- can detect nodules of
0.5cm (CXR smallest 3cm)
• High false positive rate
• Encouraging symptom awareness
• Study in Doncaster to encourage reporting
cough and early CXR increased rate of
cancers diagnosed in stage 1 or 2 from
11% to 19% after campaign
Lung cancer
• 25% lung cancers have normal CXR
• But CXR cheap and easy to obtain
• Estimate if CXR all smokers over 60 with
cough would speed the time to diagnosis
by 9 months in 25% of cancer cases
• Haemoptysis is relatively uncommon in
lung cancer- only 40% report it
• Haemoptysis risk is 2.4-7.5% depending
on age
Lung cancer
• Cough , dyspnoea , weight loss, fatigue ,
chest pain, loss of appetite commoner
symptoms than haemoptysis
• But even more common in benign
conditions- positive predictive value 1%
• UK delay in diagnosis 51 days; Sweden 26
days
• Low threshold for CXR good strategy
Colorectal cancer
• Current UK guidelines 6 weeks of bleeding or
bleeding with diarrhoea for TWR
• Over 60 risk of cancer > 2% for any bleeding so
are current guidelines correct?
• Iron deficient anaemia- men <11 women <10 –
risk of cancer 7.7% and 13.3% respectively
• Abdominal pain , constipation and diarrhoea
much lower predictive value (<1%)
• Problem is colonoscopy more expensive , some
risk, uncomfortable
Colorectal cancer
• CT colonography misses 1 in 10 lesions
larger than 1cm – which colonoscopy
picks up
• Biomarkers may help in future- matrix
metalloproteinase-9, free DNA in rectum
possibilities for future
• Risk scores – CAPER and SELVA- but
both lead to too many false positives and
excessive interventional investigation
Colorectal cancer
• GPs often do get it right
• “change in bowel habit” means diarrhoea
or constipation and reasonable chance of
bowel cancer
• Has 4% predictive value for cancer in 80yr
old , whereas constipation or diarrhoea
has 0.7% and 1.2%
• Unclear what these soft features are that
GP uses to label as change in bowel habit
Prostate cancer
• Debate re importance of early diagnosis
• Screening trials show little benefit to
mortality
• Treating small cancers gives small
mortality gain , but possible complications
; incontinence/ impotence
• Treatment of larger cancers or spread less
controversial- and these usually cause
symptoms
Prostate cancer
• Suggests definitely appropriate to search
for cancer if LUTS
• But positive predictive value only 3% for
cancer
• Impotence can be an early symptom of
cancer – rare
• Remember PSA in back pain in men
Ovarian cancer
• Symptoms are common and early- and
often missed
• Fatigue, bloating, abdominal pain, urinary
frequency
• Combination of Ca125 and TV ultrasound
probably best initial investigation
Other cancers
• Oesophageal cancer 5.7% risk in men with
dysphagia
• Urinary tract cancer 7.4% risk in men with
haematuria (women 3.4%)
• <2% lymphadenopathy presented in 1ary
care malignant
• Brain tumour risk with headache 0.1%
Tumour markers
• Over 50% tumour markers requested in
primary and secondary care inappropriate
– and requests for PSA in women and
CA125 in men not unusual!
• Key points 1/measuring more than one not
helpful (except germ cell HCG and AFP)
• 2/ not helpful in non specific symptoms –
often raised in benign disease also
Tumour markers
• Main uses in monitoring cancersresponse to treatment or detecting
recurrence
• Not helpful in screening asymptomatic
individuals
• Except AFP for hepatocellular ca in
cirrhosis from Hep B or C
• Ca125 for ovarian cancer (with TVUS) and
PSA debatable
Tumour markers
• PSA - prostate ; also elevated in BPH ,
prostatitis , UTI , cycling! PR doesnt raise
. Finasteride reduces PSA
• C125- ovary . Also raised in
endometriosis, menstruation, colitis,
laparoscopy, SLE. In conjunction with
TVUS. Can be negative despite cancer –
refer if suspicious
Tumour markers
• CEA colon cancer – only useful for
monitoring
• CA153 breast cancer – monitoring
• CA19-9 pancreatic cancer – can aid
diagnosis
• HCG and AFP germ cell tumours – can
aid diagnosis and AFP in hepatocellular
ca
Tumour markers
• Calcitonin ; medullary thyroid cancer- can
aid diagnosis
• Thyroglobulin ; follicular /papillary thyroid
cancer -can aid diagnosis
• Paraproteins eg Bence Jones – myeloma;
useful in diagnosis
• Ref BMJ 2009 ; 339:b3527
Conclusions
• GPs may have too high a threshold for
investigation
• Gatekeeper role may inhibit early
diagnosis, and financial pressures may be
increasing this
• Two week rule does not appear to have
made a significant impact on speeding
cancer diagnosis