National Breast and Ovarian Cancer Centre

• About NBOCC
• Resources
  • NBOCC News
    • Subscribe
    • Research summaries
    • Clinical Update Breast Cancer
    • Policy submissions
  • DVD
  • Hospital Services Directory
  • Other useful websites
• Media
• Events

Clinical Update - Breast Cancer

Clinical Update Issue 22 - December 2005 - ISSN 1328-9454 See previous editions

DIAGNOSTIC PERFORMANCE OF DIGITAL VERSUS FILM MAMMOGRAPHY FOR BREAST CANCER SCREENING

Commentary by Clinical Associate Professor Jonathan Osborne

The article:

Pisano ED, Gatsonis C, Hendrick E et al for the Digital Mammographic Imaging Screening Trial (DMIST) Investigators Group. Diagnostic performance of digital versus film mammography for breast-cancer screening. The New England Journal of Medicine 2005; 353(17):1773-1783

Reviewer:

Clinical Associate Professor Jonathan Osborne is the State Radiologist for BreastScreen Queensland.

In this issue...

• Article summary
• What does this article add to existing clinical evidence in this area?
• How adequate was the methodology used in addressing the aims of the study?
• What are the implications of the study for clinical practice in Australia?

Summary

Abbreviations

Area Under Receiver-Operating-Characteristic Curve (AUROCC); Confidence Interval (CI); Digital Mammographic Imaging Screening Trial (DMIST)

Study design

This is a controlled clinical trial of 49,528 asymptomatic women presenting for screening mammography at 33 sites in the USA and Canada. All women underwent both digital and film mammography. Complete results were available for 42,760 women, used for the primary analysis. Mammograms were interpreted independently by two radiologists. Breast cancer status was identified as positive or negative on the basis of a breast biopsy performed within 455 days after the initial study mammogram and/or follow-up mammogram after 1 year. Diagnostic accuracy was measured by the AUROCC (1.0 = perfect test). The primary outcome was the difference in diagnostic accuracy between digital and film mammography in the whole group. The secondary outcomes were differences in diagnostic accuracy in prespecified subgroups based on age (younger than 50 years vs. 50 years or older), breast density (heterogeneously dense or extremely dense vs. less dense), menopausal status (pre-menopausal or peri-menopausal vs. post-menopausal), race (white vs. black vs. other), risk of breast cancer (lifetime risk ≥25% vs. <25%) and the four digital-machine manufacturers.

Findings

Primary outcome:

Diagnostic accuracy of digital and film mammography was similar over the entire study population (difference in AUROCC: 0.03; 95% CI: -0.02 to 0.08; p=0.18).

Secondary outcomes:

Diagnostic accuracy of digital mammography was significantly higher than film mammography among:

• women younger than 50 years (difference in AUROCC: 0.15; 95% CI: 0.05 to 0.25; p=0.002)
• women with heterogeneously dense or extremely dense breasts on mammography (difference in AUROCC: 0.11; 95% CI: 0.04 to 0.18; p=0.003)
• pre-menopausal or peri-menopausal women (difference in AUROCC: 0.15; 95% CI: 0.05 to 0.24; p=0.002).

Diagnostic accuracy of digital mammography did not vary from that of film mammography according to race, risk of breast cancer, or type of digital machine used.

Conclusion

The authors concluded that the overall accuracy of digital and film mammography in screening for breast cancer was similar. However, digital mammography was significantly better than conventional film mammography at detecting breast cancer in young women, pre-menopausal and peri-menopausal women, and women with mammographically dense breasts.

What does this article add to existing clinical evidence in this area?

The DMIST publication has been one of the most eagerly awaited trials of 2005. Would it prove to the sceptics that digital imaging was superior to film/screen mammography? What is the best form of digital imaging – computed radiography (CR), full field, or slot scanning? These questions and others were anticipated by the DMIST study findings.

It would appear that for women in the target screening age group in Australia (50–69), digital mammography appears to offer little advantage over film/screen combination in cancer detection. This was a disappointing finding. However, the study did find statistically superior rates of cancer detection with digital mammography in three sub-groups of women – those younger than 50 years, those with dense breasts and those who were pre- or peri-menopausal. (Women with dense breasts accounted for nearly half (47%) of the women examined). In other words, cancers more likely to be missed with traditional methods because of dense breast tissue were more likely to be detected with digital mammography.

How adequate was the methodology used in addressing the aims of the study?

This is the largest study of its kind including over 40,000 women aged from 47 to 62 years.

This study did not differentiate between digital mammography technologies.  Five systems were used.  However, while the characteristics of the women screened were analysed in detail, there was no breakdown in the number of examinations performed by, or cancer detection rates accorded to, each machine type.  This information might follow in subsequent publications. 

The DMIST study did not distinguish women using Hormone Replacement Therapy (HRT).  However, it is a logical mental step to propose that if women were taking a type of HRT that increased breast density then tumours in these women would also be potentially visualised better.

What are the implications of the study for clinical practice in Australia?

This is an important study, and hopefully, remains a work in progress.  It does refer to other advantages of digital mammography, specifically the ease of image access, computer-aided detection (CAD), transmission of images and potentially lower radiation dose (still unproven).  Digital mammography is certainly feasible, and may allow improvements in workflow and workforce.  However, these are secondary reasons for change, and although important, are only relevant if digital mammography is at least as good at detecting cancer as current techniques.  Thus, from this viewpoint, the DMIST study was a success in that it showed that digital mammography was equivalent to film/screen mammography in women older than 50 years or with fatty or scattered fibro-glandular breasts and superior in women younger than 50 years, who were pre- or peri-menopausal, or had dense breasts.

Disclaimer:
Clinical Update is produced by the National Breast Cancer Centre (NBCC) and is intended to provide health professionals with timely expert commentary on new research in breast cancer. Commentaries included in Clinical Update do not replace recommendations included in NBCC clinical practice guidelines.
Information contained in Clinical Update is not intended to be used as substitute for an independent health professional's advice. The NBCC does not accept any liability for any injury, loss or damage incurred by use of or reliance on the information contained in Clinical Update. The NBCC develops material based on the best available evidence however cannot guarantee and assumes no legal liability or responsibility for the currency or completeness of the information.

Back to Top