Table J5 - Screening for women with a family history of breast cancer

A Evidence based guidelines - the evidence base for mammographic screening of women with a family history of breast cancer

Evidence Table J5

Screening for women with a family history of Breast Cancer

A	Clinical Problem	Screening for women with a family history of breast cancer
A1	Ref No. in MBUR4 (if any)	J5
A2	Section in MBUR4	Breast disease
B	Search strategy: databases used; period; MeSH headings; other key words.	Medline 1966- June 2001, EMBASE 1988-2001, diagnostic imaging, breast neoplasms, family history
C	Search results: no. found; no. used.	104 found, 7 used
D	Reference numbers of cited references in attached master list.	J5 1-7
E	Summarised results from each of the cited references	Kerliskowe et al (J5 1) report the largest scale study so far of women at increased risk because of family history in a cross sectional study of 389,533 screened women comparing results in those with and without a first degree relative with breast cancer and found results of screening in high risk women to be similar to those in the general population who are a decade older. Tilanus-Linthorst (J52), Moller (J5 5), Laloo (J5 6) and Kollias (J5 7 ) all report smaller cohort studies of women with family histories of breast cancer under age 50 showing similar results of screening to findings in over 50 year olds in National and large scale screening programmes. Moller (J5 5) reports 5 year survival from mammographic screen detected breast cancer in high risk younger women comparable with that from screen detected cancer in older women. Kuhl (J5 3) compares the use of mammography, US and MRI screening in gene carriers. MRI has the highest sensitivity. Eccles summarises guidelines from the UK Family Study Group (J5 4).
F	Statement (= the conclusion drawn from E)	Evidence is now emerging that the results of screening younger women (most studies average age between 40 and 50) at high risk because of a family history of breast cancer are comparable to results of National programmes and RCTs of population screening in over 50 year olds in terms of number and stage of cancers detected. Evidence for mortality reduction is limited because of the small size of these studies compared with RCTs of population screening. There is ongoing research, much of it directed at screening modalities other than mammography as the sensitivity of mammography is less in younger women. There is consensus that the benefit of screening outweighs the risks in younger women who are at significantly increased risk. Mammographic screening should be offered to them, after careful assessment of their level of risk and counselling about the benefits and drawbacks of screening under the auspices of specialist family history clinics.
G	Evidence level of F classified I - IV	2
H	Investigation	Mammography
I	Recommendation *	Specialised examination
J	Grade of Recommendation A - C	B
K	Comment (if any) to go in booklet version of MBUR5	Evidence of benefit is emerging for women at significantly increased risk in their 40s and appears to outweigh the harm of screening. Screening should only be undertaken after genetic risk assessments and appropriate counselling as to the risks and benefits. Consensus is that screening women under age 50 with a family history should only be undertaken when the lifetime risk of breast cancer is greater than 2 times the average. Further guidelines for mammographic and other forms of screening in these women remain under review.
L	Any other comments on this problem (e.g. caveats; suggestions for research or systematic review; cost or opportunity cost; users' views.	This is a thorny topic and would benefit from a systematic review as there have been a number of new publications recently. Ros Given-Wilson, Erika Denton

Comments: Family history screening is best performed in conjunction with the cancer genetics service according to the highest standards available - ie the NHS BSP/Breast Test Wales/Scottish BSP

References: -

Ref ID	Author; Year; Country; Grade (I-IV).	Aims	Patient population	Study design*	Results
J5 1	Kerliskowe K. 2000 USA 2	Performance of screening in women with and without a family history	389,533 women age 30-69 undergoing breast screening	Large scale cross sectional study	Results of screening in women with a first degree relative with breast cancer were comparable to those without who were a decade older. Sensitivity of mammography increased with age. The PPV for screening was greater in high risk women.
J5 2	Tilanus-Linthorst MM. 2000 Netherlands 3	Assess stage of screen detected cancers compared with symptomatic cancers in family history women.	678 screened women who were at increased risk. Also MRI screening in a subgroup who had dense breasts.	Cohort	26 screen detected breast cancers were significantly lower stage than symptomatic ones and comparable to the stage of screen detected cancers form population screening. MRI found 3 occult cancers.
J5 3	Kuhl CK. 2000 Germany 3	To compare MRI with Mammography and US in screening unsuspected or proven gene carriers	192 asymptomatic and 6 symptomatic potential gene carriers	Cohort	6 cancers were found in the symptomatic patients and 9 in the asymptomatic. In the asymptomatic group sensitivities were MRI 100%, Mammo 33%, US 33%.
J5 4	Eccles DM. 2000 UK 4	Guidelines from Family Study Group	N/A	Experts opinion	Guidelines
J5 5	Moller P. 1999 Norway 3	Assess stage and 5 year survival of women with FH screen detected cancers	161 women undergoing FH screening in 7 centres with breast cancer	Multicentre cohort	75% of cancers were screen detected. Overall 5 year survival was 89% (100%) for women with DCIS.
J5 6	Lalloo F. 1998 UK 3	Report results of FH screening women <50	1259 women <50 with a 1 in 6 or greater lifetime risk of breast cancer	Cohort	Cancer detection rates similar to those in older women undergoing population screening
J5 7	Kollias J. 1998 UK 3	Report results of FH screening women < 50	1371 women < 50 with a FH of BC	Cohort	Detection rates similar to NHSBSP. More DCIS in the FH group but similar stage of invasive cancer to the symptomatic population.