Breast cancer screening in British Columbia: A guide to discussion with patients - BC Cancer
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Colin Mar, MD, FRCPC, Janette Sam, RTR, Christine Wilson, MD, FRCPC
Breast cancer screening in
British Columbia: A guide to
discussion with patients
Primary care providers have an important role to play in helping
their patients consider the benefits, limitations, and downsides of
screening mammography.
T
ABSTRACT: Breast cancer contin- recommended for women age 40 to he potential impact of breast
ues to affect the women of British 74 with a first-degree relative with cancer on women in British
Columbia and impose a significant breast cancer. The Screening Mam- Columbia means that primary
health care burden. Population- mography Program compiles data care providers should be prepared to
based screening mammography for calculating numerous outcomes, address the female patient’s question,
remains the most accessible and including participation and return “Should I get a mammogram?” It can
scientifically validated test for de- rates, time to diagnosis measures, be helpful to begin with a brief review
tecting breast cancer and reducing and sensitivity and specificity indi- of some principles of screening from
breast cancer mortality. Screening is cators. Breast density is an issue a the classic World Health Organization
provided across the province by the woman and her primary care provid- report by Wilson and Jungner:
Screening Mammography Program er may need to consider, since nor- • The condition sought should be an
of BC. Downsides of screening in- mal dense breast tissue may impede important health problem.
clude exposure to ionizing radiation, detection of cancer. Imaging tech- • There should be an available
false-positive results, and overdiag- nologies undergoing investigation to treatment.
nosis. Current screening policy in BC address this and other challenges in- • There should be an acceptable test.1
is based on age and other determi- clude digital breast tomosynthesis, These principles have undergone
nants of risk, including family histo- ultrasound, and magnetic resonance multiple revisions over the years and
ry and genetic factors. For example, imaging (MRI). By discussing imag- additional criteria have been pro-
routine screening every 2 years is ing options and screening benefits, posed, including:
recommended for asymptomatic limitations, and downsides with • There should be scientific evidence
women age 50 to 74 of average risk, women, primary care providers can
while routine screening every year is facilitate informed decision making. Dr Mar is the medical director of the Screen-
ing Mammography Program of BC and a
clinical assistant professor in the Depart-
ment of Radiology at UBC. Ms Sam is the
operations director of the Screening Mam-
mography Program of BC. Dr Wilson is the
director of breast imaging at BC Cancer in
Vancouver and a clinical associate profes-
sor in the Department of Radiology at UBC.
This article has been peer reviewed.
20 bc medical journal vol. 60 no. 1, january/february 2018 bcmj.orgBreast cancer screening in British Columbia: A guide to discussion with patients
of screening program effectiveness. intention-to-treat analysis. Mammog- al screening for women 45 to 54. The
• There should be quality assurance. raphy technology has evolved since ACS update also included qualified
• The overall benefits of screening 2000, and quality assurance programs recommendations for biennial screen-
should outweigh the harm.2 have been developed. Observational ing beginning at age 55, and contin-
Although breast cancer screening, studies have yielded more recent data. ued screening at 70 to 74 to be based
treatment, and surveillance in BC These include the work of Coldman on life expectancy. Finally, the update
have contributed to outcomes match- and colleagues, who considered over included a qualified recommendation
ing national and international stan- 2 million women age 40 to 79 in 7 that women “should have the oppor-
dards,3 the disease remains the most of 12 Canadian screening programs tunity to begin annual screening be-
common cancer in women in Canada, during the period 1990 to 2009,9 and tween the ages of 40 and 44 years.”
and the second leading cause of can- observed a 40% mortality reduction, Since a qualified recommendation
cer death as of 2015.4 In 2010 the life- with little variation by age. The num- indicates “there is clear evidence of
time risk for developing breast cancer ber needed to participate in screening benefit, but less certainty about either
was 1 in 9 and the lifetime mortality to prevent a single breast cancer death the balance of benefits and harms,
risk was 1 in 30. Since 1986 there within 10 years decreased with age or about patients’ values and prefer-
has been a steady decline in the age- from 1247 for women first screened at ences,”11 different patients offered
standardized breast cancer mortal- age 40 to 49, to 498 for women first this opportunity will make differ-
ity rate, which now stands at 16 per screened at age 70 to 79. ent decisions and discussion will be
100 000 in BC. The age-standardized The Canadian Task Force on Pre- required. In these cases, the primary
5-year relative survival ratio for 2006 ventive Health Care (CTFPHC) last care provider has an important role to
to 2008 was 88% across the country.4 issued guidelines for breast cancer play in facilitating informed decision
Population screening for breast screening in 2011.7 These included a making.
cancer in BC began in 1988 with the weak recommendation for mammog- None of the three organizations
Screening Mammography Program raphy every 2 to 3 years for women has found sufficient evidence to sup-
(SMP). There are SMP centres locat- age 50 to 69, and the same for women port a recommendation for routine
ed in all five health authorities, with 70 to 74. Evidence of similar quality clinical breast examination or breast
36 fixed sites across the province and supporting a weak recommendation self-exam. The American Cancer So-
three mobile units providing access for mammography for women age ciety has suggested that the time re-
for remote and underserviced regions. 40 to 49 was reported, but a recom- quired for clinical breast examination
The program is completing a transi- mendation was not provided after the instead be used for discussion of the
tion to digital mammography, which CTFPHC cited a less favorable bene- benefits, limitations, and downsides
facilitates the transfer of images be- fit-to-harm ratio in this age group. of mammography.
tween centres, and has been shown to A working group of the Interna-
improve cancer detection in younger tional Agency for Research on Can- Downsides of screening
women and those with dense breast cer (IARC) subsequently published mammography
tissue.5 an evidence review in 2015.10 They As noted previously, a discussion
concurred with the CTFPHC in that of screening requires considering
Evidence for screening they found sufficient evidence to rec- downsides. These include exposure
mammography ommend screening for the 50 to 69 to ionizing radiation, patient anxiety,
The evidence for breast cancer screen- and 70 to 74 age groups. They were, false-positives, and overdiagnosis.
ing with mammography has en- however, unable to find sufficient evi- The radiation risk posed by current
gendered much discussion. Several dence to make a recommendation for standards in digital mammography is
randomized controlled trials were women 40 to 49, citing fewer studies low. For a woman undergoing mam-
conducted prior to 2000 and meta- for this age group. mography at age 40, the estimated life-
analyses of these demonstrated re- The American Cancer Society time attributable risk (LAR) of a fatal
ductions in breast cancer mortality (ACS) released a guideline update breast cancer is 1.3 cases per 100 000.
with screening of RR 0.80 to 0.82.6-8 in 201511 that included a strong rec- Continuation with annual mammog-
These studies may, however, under- ommendation for regular screening raphy to the age of 80 is associated
estimate the current effectiveness of mammography starting at age 45 and with an LAR of 20 to 25 cases.12 The
screening given their age and use of a qualified recommendation for annu- IARC included a statement in its 2015
bc medical journal vol. 60 no. 1, january/february 2018 bcmj.org 21Breast cancer screening in British Columbia: A guide to discussion with patients
guidelines that the risk of radiation- BC screening policy in older age groups, is the greater
induced malignancy is outweighed by Population screening recommen- prevalence of dense breast tissue that
the benefits of mammography.10 dations for breast cancer in BC are may impede cancer detection.20
False-positives in screening mam- categorized by age and other deter- For women at high risk due to a
mography are inherent to the practice. minants of risk, including family genetic predisposition or a history
In 2015 the positive predictive value history and genetic factors ( Table ). of chest wall irradiation between the
in the SMP for first screens was 5.2% Women eligible for screening are ages of 10 and 30 years, screening
and for subsequent screens was 7.0%, asymptomatic, without a personal MRI is recommended in addition to
values that both met national targets.13 history of breast cancer, and with- annual mammography, although MRI
Anxiety caused by false-positives is out breast implants. Women age 40 is not provided through the SMP.
related to receiving notification of to 49 are encouraged to consider the Regarding clinical breast exami-
an abnormal result and undergoing benefits relative to the downsides and nation, there is no recommendation for
consequent image-guided or surgi- limitations in discussion with their or against this practice in asymptom-
cal biopsy. Such anxiety has been primary care provider. A limitation atic women. Finally, the policy recom-
documented,14 but has not been found for women in this age group, where mends against breast self-examination
to have a measurable health utility the incidence of cancer is lower than as an alternative to mammography.
decrement.15 There are also varying
morbidity and risks associated with Table. Screening Mammography Program of BC guidelines for primary care providers.
different biopsy procedures. Over-
diagnosis involves the detection by
mammography of a malignancy that
would never have become clinically
apparent before the patient’s death.
The issue then is one of results that
precipitate overtreatment. The mea-
surement of this is complicated, pri-
marily by uncertainty regarding the
true incidence of breast cancer, the
subject of much discussion and de-
bate. Recently published overdiagno-
sis rates range from 2.3% in a Danish
population-based cohort study16 to
48.3% in a later Danish study that in-
cluded findings for invasive cancer
and ductal carcinoma in situ (DCIS).17
A retrospective study of provincial
SMP data estimated rates of 5.4%
for invasive cancer and 17.3% when
DCIS was included, with the risk of
overdiagnosis being highest in older
women.18 In discussing this issue with
patients it remains important to recog-
nize that the lifetime risk for overdi-
agnosis is low, in the order of 1.0%.19
Moreover, it is not currently possible
at the time of diagnosis to distinguish
between tumors that will not progress
from those that will. The decision will
therefore be based on the patient’s tol-
erance of the relative risks. Source: BC Cancer
22 bc medical journal vol. 60 no. 1, january/february 2018 bcmj.orgBreast cancer screening in British Columbia: A guide to discussion with patients
255 534 screens
Normal Abnormal
232 382 (91% of total) 23 152 (9% of total)
Benign/normal on imaging workup Further diagnostic workup Insufficient follow-up procedure
18 871 4 109 information
(82% of those with follow-up) (18% of those with follow-up) 172 (1% of abnormal)
Diagnosis at core/Fine needle aspiration Diagnosis at open biopsy
3 406 703
(83% of further diagnostic work-up) (17% of further diagnostic work-up)
Benign Malignant Benign Malignant
2 150 (63% of core/ 1 256 (37% of core/ 551 152
fine needle aspiration) fine needle aspiration) (78% of open biopsy) (22% of open biopsy)
Ductal carcinoma in situ Invasive Ductal carcinoma in situ Invasive
221 1 035 86 66
(18% of malignant) (82% of malignant) (57% of malignant) (43% of malignant)
Figure 1. Breast cancer screening outcomes, 2015.13
Source: BC Cancer
Screening program participation by region and by select- lies above the provincial average.
outcomes ed ethnic groups. This interpretation may, however, be
Objective outcome measures are inte- In 2015 the provincial participa- limited by underestimation of the eth-
gral to quality assurance in a screen- tion rate for women age 50 to 69 was nic group populations.13
ing program. The SMP compiles data 52.4%, a rate that has remained both More than 250 000 mammograms
for calculating numerous outcome relatively stable and below the na- were performed by the SMP in 2015.
measures that are then shared in a var- tional target of 70.0% since 2000.13 Screening outcomes considered in-
iety of ways. The program’s annual Participation rates by women age 50 cluded normal and abnormal results,
report is the most comprehensive of to 69 in the Northeast health service image-guided and surgical biopsies
these, and may be accessed at www.bc delivery area (40.0%) and Kootenay performed, and breast cancers de-
cancer.bc.ca/screening/Documents/ Boundary health service delivery area tected ( Figure 1 ) . The percentage
SMP_Report-AnnualReport2016 (44.0%) were below the provincial of women referred for further test-
.pdf. The report includes participa- average. Data were compiled for cli- ing because of an abnormal screen-
tion and return rates, time to diagnosis ents identified as First Nations, East/ ing mammogram (i.e., the abnormal
measures, and sensitivity and speci- South East Asian, and South Asian. call rate) was 9.1%. The number of
ficity indicators, and compares these The participation of women within women with a screen-detected cancer
indicators to national standards where the same age range in all three groups per 1000 women who had a screening
available. The program also considers rose over the previous 5 years, and mammogram (i.e., the cancer detec-
bc medical journal vol. 60 no. 1, january/february 2018 bcmj.org 23Breast cancer screening in British Columbia: A guide to discussion with patients
tion rate) was 5.5. The percentage of Breast density 226 000 controls to determine a 4.64-
women with an abnormal mammo- Risk stratification in the current SMP fold risk when the proportion of dense
gram who were diagnosed with breast screening policy is based primar- to non-dense breast tissue was equal
cancer (i.e., the positive predictive ily on age and family history, but to or greater than 75%,21 relative to
value) was 6.1%.13 another factor a woman and her pri- a breast of less than 5%. Breast can-
In addition to the outcome mea- mary care provider should consider is cer in the setting of dense breast tis-
sures already noted, individualized breast density. This is a measurement sue has not, however, been associated
data are compiled for each radiolo- of the proportion of the breast com- with an increased risk of death.22
gist screener in the program and for posed of dense (i.e., nonfatty) tissue, The reporting of breast density
each provincial health authority. The and the probability of masking a can- with mammogram results has been
SMP also promotes quality assurance cer. Dense tissue is relatively radio- the focus of much recent discussion.
through a client satisfaction survey opaque and thus appears white on a As of early 2016, 24 American states
sent to selected women attending pro- mammogram. Although normal, such have enacted legislation that man-
gram sites across the province. Each tissue may obscure cancer and thus date this reporting.23 While there is no
SMP site and radiologist maintains impede its detection.20 A set of mam- legislative requirement in Canada to
mammography-specific accredita- mograms ( Figure 2 ) illustrates the report on breast density, the SMP pol-
tion from the Canadian Association difference between dense and non- icy on reporting of breast density is
of Radiologists. This requires adher- dense breasts, and how cancer may currently under review. For now, the
ence to a nationally recognized set of resemble dense tissue. Given this information is available upon patient
guidelines that ensures the quality of masking effect, any breast changes or request, with the understanding that
the examination and the competence symptoms should be followed up, re- this risk factor should not be consid-
of the screener. Finally, specific poli- gardless of a normal screening mam- ered in isolation, but in combination
cies regulate the systematic review of mogram result. with age, family history, and other
randomly sampled abnormal findings Breast density is also a risk fac- risk factors.24 At this time, there have
and cancers diagnosed. This occurs at tor for incident cancer, particularly been no guideline revisions regarding
both the site level to facilitate direct when the breast is extremely dense. supplemental screening for women
feedback and at the program level to A meta-analysis in 2006 considered with dense breasts.
ensure overall effectiveness. over 14 000 breast cancer cases with
A B C
Figure 2. Mammograms illustrating the challenge breast density may pose in their interpretation. A: Low opacity seen in non-dense breast
with a high proportion of fat. B: Increased opacity seen in a dense breast. C: Opacity seen in both normal dense breast tissue (solid arrow) and
adjacent cancer (dashed arrow).
24 bc medical journal vol. 60 no. 1, january/february 2018 bcmj.orgBreast cancer screening in British Columbia: A guide to discussion with patients
Emerging technologies the use of MRI for high-risk women sponse “and discuss with your doctor
Mammography remains the most ac- with genetic or familial risks or prior to determine if screening is right for
cessible and scientifically validated mantle radiation exposure. MRI has you.” Research within the program is
test for breast cancer screening, and demonstrated high sensitivity for de- now underway to consider the roles
the sole modality included in guide- tecting breast cancer, but the use of of both breast density and ethnicity
lines for women of average risk. It this modality is limited by examina- within BC and Canada, and this may
is, however, helpful to have a basic tion time and geographic availability. further individualize the assessment
understanding of some of the other Recent evaluation of abbreviated im- of risk.
breast imaging modalities that are the aging protocols for this modality may,
focus of ongoing research, and may however, eventually allow its use for Summary
arise in discussion with patients. other risk categories by increasing ac- Both population-based screening and
Digital breast tomosynthesis cess through shortened time required treatment advances have improved
(DBT) is commonly referred to as the per visit.28 breast cancer outcomes. This dis-
3-D mammogram. Indeed, this ex- Other imaging modalities such as ease, however, continues to impose
amination utilizes the technology of thermography and nuclear medicine a significant burden on the health of
mammography to produce a series of tests, including positron emission to- women across Canada.4 Mammog-
two-dimensional images of a single mography, have not been validated raphy remains the most scientific-
breast. These are acquired through an for population screening. ally validated screening test to reduce
arc trajectory, and ultimately viewed breast cancer mortality. A woman’s
as a three-dimensional image set. Facilitating an participation in a mammography
Multiple studies have demonstrated informed decision screening program is best predicat-
the ability of DBT to increase cancer By increasing awareness of risk ed on an informed decision. This re-
detection while decreasing the rate factors for breast cancer, the SMP quires considering risk factors and
of patient recall for further evalua- hopes to help women in BC age 40 understanding the limitations and
tion.25 A prospective trial integrating and older make an informed deci- downsides of screening.
2-D and 3-D mammography to screen sion about screening mammography. We encourage the public to use
over 7000 women in 2013 found an When researchers analyzed provin- the Breast Cancer Screening Decision
additional 2.7 cancers were detected cial screening data for over 2 million Aid discussed above and to access
per 1000 screens, and false-positives women age 40 to 74 screened be- resources for general information on
were reduced by an estimated tween 2000 and 2009, they found de- screening (www.screeningbc.ca). We
17.2%.26 Two sites within the SMP creased false-positives and increased also encourage primary care provid-
are currently participating in a large cancer detection with increasing age, ers to take advantage of the continuing
multicentre trial to evaluate the role and increased cancer detection with a professional development resources
of DBT in screening. positive family history. The main fac- available (http://ubccpd.ca/course/
Ultrasound is integral to breast tors associated with false-positives bca-screening-update).
imaging in the diagnostic setting, were time since last screening and a Breast cancer screening policy in
and its role in screening is evolving. previous false-positive.29 These and BC will continue to evolve through
It is of particular interest in the con- other findings were used to develop ongoing internal data analysis, ap-
text of dense breast tissue. An earlier the online Breast Cancer Screening praisal of the medical literature and
prospective multicentre trial followed Decision Aid of BC Cancer (http:// review of working group guidelines
women over three rounds of annual decisionaid.screeningbc.ca), which that address risk factors such as breast
mammography with and without sup- generates a response after a user an- density, and the development of other
plemental ultrasound. Inclusion cri- swers six questions, including “How breast imaging modalities.
teria were a breast density of at least old are you?,” and “When was your
50% and at least one other risk factor, last screening?” The response indi- Competing interests
such as a personal history of breast cates the likelihood of three events: None declared.
cancer. An additional 3.7 cancers per having a breast cancer found, hav-
1000 screens were detected, but with ing a false-positive mammogram, and References
a false-positive rate of 16%.27 having a false-positive biopsy. The 1. Wilson JMG, Jungner G. Principles and
SMP screening policy indicates user is then advised to print the re- practice of screening for disease. Geneva:
bc medical journal vol. 60 no. 1, january/february 2018 bcmj.org 25Breast cancer screening in British Columbia: A guide to discussion with patients
WHO; 1968. Accessed 27 February 2017. average risk: 2015 guideline update from et al. Relationship between mammo-
http://apps.who.int/iris/bitstream/ the American Cancer Society. JAMA graphic density and breast cancer death in
10665/37650/17/WHO_PHP_34.pdf. 2015;314:1599-1614. the Breast Cancer Surveillance Consor-
2. Andermann A, Blancquaert I, Beauchamp 12. Hendrick RE. Radiation doses and cancer tium. J Natl Cancer Inst 2012;104:1218-
S, Déry V. Revisiting Wilson and Jungner risks from breast imaging studies. Radiol- 1227.
in the genomic age: A review of screening ogy 2010;257:246-253. 23. Destounis S, Johnston L, Highnam R,
criteria over the past 40 years. Bull World 13. Screening Mammography Program, BC et al. Using volumetric breast density to
Health Org 2008;86:241-320. Cancer Agency. Screening Mammog- quantify the potential masking risk of
3. Coleman MP, Forman D, Bryant H, et al. raphy Program 2016 annual report. Ac- mammographic density. AJR Am J Roent-
Cancer survival in Australia, Canada, Den- cessed 19 October 2017. www.bccancer genol 2017;208:222-227.
mark, Norway, Sweden, and the UK, .bc.ca/screening/Documents/SMP 24. Kerlikowske K, Zhu W, Tosteson AN, et
1995–2007 (the International Cancer _Report-AnnualReport2016.pdf. al.; Breast Cancer Surveillance Consor-
Benchmarking Partnership): An analysis 14. Miller LS, Shelby RA, Balmadrid MH, et al. tium. Identifying women with dense
of population-based cancer registry data. Patient anxiety before and immediately breasts at high risk for interval cancer: A
Lancet 2011;377(9760):127-138. after imaging-guided breast biopsy pro- cohort study. Ann Intern Med 2015;
4. Canadian Cancer Society’s Advisory Com- cedures: Impact of radiologist-patient 162:673-681.
mittee on Cancer Statistics. Canadian communication. J Am Coll Radiol 2013; 25. Hooley RJ, Durand MA, Philpotts LE. Ad-
Cancer Statistics 2015. Toronto, ON: Ca- 10:423-431. vances in digital breast tomosynthesis.
nadian Cancer Society; 2015. 15. Tosteson AN, Fryback DG, Hammond CS, AJR Am J Roentgenol 2017;208:256-266.
5. Pisano ED, Hendrick RE, Yaffe MJ, et al.; et al. Consequences of false-positive 26. Ciatto S, Houssami N, Bernardi D, et al.
DMIST Investigators Group. Diagnostic screening mammograms. JAMA Intern Integration of 3D digital mammography
accuracy of digital versus film mammog- Med 2014;174:954-961. with tomosynthesis for population breast-
raphy: Exploratory analysis of selected 16. Njor SH, Olsen AH, Blichert-Toft M, et al. cancer screening (STORM): A prospec-
population subgroups in DMIST. Radiolo- Overdiagnosis in screening mammog- tive comparison study. Lancet Oncol
gy 2008;246:376-383. raphy in Denmark: Population based co- 2013;14:583-589.
6. Independent UK Panel on Breast Cancer hort study. BMJ 2013;346:f1064. 27. Berg WA, Zhang Z, Lehrer D, et al. Detec-
Screening. The benefits and harms of 17. Jørgensen KJ, Gøtzsche PC, Kalager M, tion of breast cancer with addition of an-
breast cancer screening: An independent et al. Breast cancer screening in Denmark: nual screening ultrasound or a single
review. Lancet 2012;380(9855):1778- A cohort study of tumor size and overdi- screening MRI to mammography in
1786. agnosis. Ann Intern Med 2017;166: women with elevated breast cancer risk.
7. Tonelli M, Connor Gorber S, Joffres M, et 313-323. JAMA 2012;307:1394-1404.
al.; Canadian Task Force on Preventive 18. Coldman A, Phillips N. Incidence of breast 28. Chhor CM, Mercado CL. Abbreviated
Health Care. Recommendations on cancer and estimates of overdiagnosis MRI protocols: Wave of the future for
screening for breast cancer in average-risk after the initiation of a population-based breast cancer screening. AJR Am J Roent-
women aged 40-74 years. CMAJ 2011; mammography screening program. genol 2017;208:284-289.
183:1991-2001. CMAJ 2013;185:E492-498. 29. Coldman A, Phillips N, Wilson C, et al.
8. Gøtzsche PC, Jørgensen KJ. Screening 19. Marmot MG, Altman DG, Cameron DA, Information for physicians discussing
for breast cancer with mammography. et al. The benefits and harms of breast breast cancer screening with patients.
Cochrane Database Syst Rev 2013;(6): cancer screening: An independent re- BCMJ 2013;55:420-428.
CD001877. view. Br J Cancer 2013;108:2205-2240.
9. Coldman A, Phillips N, Wilson C, et al. Pan- 20. Boyd NF, Guo H, Martin LJ, et al. Mam-
Canadian study of mammography screen- mographic density and the risk and detec-
ing and mortality from breast cancer. J tion of breast cancer. N Engl J Med 2007;
Natl Cancer Inst 2014;106:dju261. 356:227-236.
10. Lauby-Secretan B, Scoccianti C, Loomis 21. McCormack VA, dos Santos Silva I. Breast
D, et al. Breast-Cancer Screening—view- density and parenchymal patterns as
point of the IARC Working Group. N Engl markers of breast cancer risk: A meta-
J Med 2015;372:24. analysis. Cancer Epidemiol Biomarkers
11. Oeffinger KC, Fontham ET, Etzioni R, et al. Prev 2006;15:1159-1169.
Breast cancer screening for women at 22. Gierarch GL, Ichikawa L, Kerlikowske K,
26 bc medical journal vol. 60 no. 1, january/february 2018 bcmj.orgYou can also read
