Hepatocellular Carcinoma: to screen or not

1
Hepatocellular Carcinoma to screen or not ?

• General Practice Unit
• Department of Medicine
• The University of Hong Kong

2
Case One

• 35 y.o. woman with known hepatitis B carrier
  status since birth of first son 12 years ago.
• She is otherwise well and is not on any
  medications.
• P/E NAD.
• She has heard from a radio programme that she is
  at risk of developing serious complications,
  including liver cancer. She wants to know if
  there are any precautionary measures she could
  take?

3
Case Two

• 55 y.o. man with known history of hepatitis for
  several years.
• He was diagnosed to have inoperable
  hepatocellular carcinoma when he developed
  jaundice in June 1995 and died 6/12 later.
• He was seen regularly at a major hospital for
  impaired liver function before HCC diagnosis.
• His wife came in to see you wanting to know
• if anything they could have done to prevent the
  cancer.

4
The object of screening for disease is to
discover those among the apparently well who are
in fact suffering from disease.

• Principles and Practice of Screening for Disease
• J.M.G. Wilson
• World Health Organization

5
Wilsons principles of early disease detection
6
1. The condition sought should be an important
health problem.

• HCC is one of the ten commonest cancers in the
  world.
• In 1992, liver cancer accounted for 9.1 of all
  newly diagnosed cases and 12.8 of all cancer
  deaths in Hong Kong.
• It ranked 2nd for males and 5th for females for
  new cancers. It also ranked 2nd for males and
  4th for females for cancer deaths.
• In Hong Kong, hepatitis B is the cause of 84 of
  HCC. 11 of males and 6.1 of females are
  carriers.

7
2. There should be an accepted treatment for
patients with recognized disease.

• No patients with large, unresectable tumors
  survive longer than five years. (Regan 1989)
• Median survival of symptomatic HCC is only 4/12.
  (Lai et al 1998)
• Patients with tumors less than 3 cm in diameter
  have a 50 five-year survival rate after
  successful surgical resection. (Iwatsuk et al.
  1983)

8
3. Facilities for diagnosis and treatment should
be available.

• Most commonly used tests alpha-fetoprotein and
  ultrasound are widely available.
• Resection of tumor i.e. hepatic resection is
  performed in many major hospitals in Hong Kong.

9
4. There should be a recognizable latent or
early symptomatic stage.

• It has been estimated that an asymptomatic tumor
  of lt5cm in diameter takes a median of 6-9 months
  to become clinically symptomatic, usually with a
  diameter of over 8 cm. (Sheu et al 1985)

10
5. There should be a suitable test or
examination.

• Sensitivity() 4/12 6/12 12/12
• alpha-fetoprotein 80 75 68
• ultrasound 98 95 88
• both 100 98 90
• (Kang et al, 1992)

11
6. The test(s) should be acceptable to the
population.

• alpha-fetoprotein a relatively non-invasive
  blood test and results available quickly.
• ultra-sound study a non-invasive test with
  little discomfort and results available
  immediately.

12
7. The natural history of the condition,
including development from latent to declared
disease, should be adequately understood.

• Hepatitis B is the cause of HCC in 84 of
  patients in Hong Kong (Lai et al 1998)
• Patients with asymptomatic HCC had a higher
  operability rate (76.5) and all had a tumor of
  less than 5 cm in diameter and had a better
  long-term survival. (Lai et al 1992)
• Symptomatic HCC 4.7 operable and only 2 out of
  191 patients had a tumor diameter of lt5cm. (Lai
  et al 1992)

13
8. There should be an agreed policy on whom to
treat as patients.

• HCC a frank disease, therefore not a problem
• Other conditions borderline condition may exist
  e.g. impaired glucose tolerance. Preventive
  management may need to be considered.

14
9. Case-finding should be a continuing process
and not a once and for all project.

• single-occasion drives would lead to only a
  small proportion (often those at least risk) to
  be screened.
• it picks up those persons in the population who
  happen at that particular time to have the
  condition but it will not affect the future
  incidence of disease.

15
10. the cost of case-finding should be
economically balanced in relation to possible
expenditure on medical care as a whole.
16
The Effectiveness of Screening

• Effectiveness of detecting HCC lt 3cm
• sensitivity of screening (66 - 100)
• Increase in survival 50-5 45
• Number needed to screen to detect one HCC
• 1000 / (prevalence rate x sensitivity of test)
• Number needed to screen to save one life
• 1000 / (mortality rate x sensitivity of
  screening x increase in survival)
• rates are per Thousand

17
Incidence Mortality Rate of HCC Among HB
Carriers

• Incidence (prevalence) of HCC
• male (1210x.84)/(1618900 x.11) 5.70/1000
• female 362 x .84/(1551500 x.06) 3.26/1000
• HCC Mortality rate
• male (888 x .84)/(1618900 x.11) 4.18/1000
• female (269 x.84)/(1551500x.06) 2.43/1000

18
Annual Cost of Screening

• Screening Cost per person per year
• APFUSS 4/12 750 x 3 2250
• USS yearly 500
• USS 6/12 500 x 2 1000
• AFP yearly 250
• AFP 6/12 250 x 2 500
• Cost per HCC detected /life saved
• screening cost per person per year x
• number of persons needed to screen per year

19
Cost-effectiveness of HCC Detection in Male HB
Carriers
20
Cost-effectiveness of HCC Detection in Female HB
Carriers
21
Cost-effectiveness of Lives Saved in Male HB
Carriers
22
Cost-effectiveness of Lives Saved in Female HB
Carriers
23
Conclusions

• Yearly AFP is the cheapest
• Yearly USS is probably the most cost-effective
  balancing the cost per life saved against cancers
  missed, especially for males
• 6/12 AFP is inferior to yearly USS both in
  cost-effectiveness and number of deaths prevented
• HCC screening is more cost-effective than breast
  cancer screening (6.6 millions per life saved).

24
Cost-effectiveness Analysis

• Prevalence (incidence)
• Sensitivity of the test
• Cost of screening
• Mortality rate
• Mortality reduction ( relative risk)