Long-term Impact of Adjunctive Ultrasonography on Advanced Breast Cancer: A Secondary Analysis of the J-START Trial

Highlights

• Adjunctive ultrasonography (US) significantly reduced the cumulative incidence of advanced breast cancer (Stage II or higher) in women aged 40–49 compared to mammography (MG) alone.
• The hazard ratio (HR) for advanced cancer was 0.83 (95.6% CI 0.70–0.98, p=0.026) over a median follow-up of 11.4 years.
• The benefit of US was most pronounced between 48 and 96 months after the initial screening, suggesting a long-term ‘stage-shift’ effect that persists well beyond the intervention period.
• These findings provide robust evidence for integrating US into screening programs for populations with a high prevalence of dense breast tissue, such as Asian women.

Background

Breast cancer remains a leading cause of morbidity and mortality globally, with significant disparities in outcomes linked to early detection capabilities. In women aged 40–49, particularly in Asian populations, traditional mammography (MG) faces a significant diagnostic hurdle: dense breast tissue. High breast density (BI-RADS categories c and d) reduces the sensitivity of MG due to the ‘masking effect,’ where radiopaque fibroglandular tissue obscures small, potentially aggressive tumors. This diagnostic gap often results in delayed diagnoses and a higher proportion of advanced-stage presentations.

The Japan Strategic Anti-cancer Randomised Trial (J-START) was initiated to address this unmet need. While the initial reports from J-START confirmed that adding ultrasonography (US) to MG increased the sensitivity and detection rate of breast cancer, the long-term clinical utility—specifically whether this increased detection translates into a reduction in advanced-stage cancers—remained the critical question for health policy and guideline development. This report summarizes a prespecified secondary analysis evaluating the 11-year cumulative incidence of advanced breast cancer.

Key Content

The Global Burden and ‘Treatable’ Deaths

To understand the significance of J-START, it is essential to frame it within the global context of cancer mortality. Recent population-based studies using GLOBOCAN data indicate that approximately 47.6% of cancer deaths globally are avoidable through primary prevention or early detection and treatment. Female breast cancer, specifically, accounts for the highest number of ‘treatable’ deaths—estimated at 0.2 million globally. The disparity is particularly sharp in countries with lower Human Development Index (HDI) scores, where access to early detection is limited. J-START represents a high-quality effort in a high-HDI setting to refine screening technology to further reduce these avoidable deaths by moving the needle from detection to stage-reduction.

Study Design and Patient Population

The J-START trial was a massive undertaking, enrolling 72,661 asymptomatic women aged 40–49 across 42 sites in Japan between 2007 and 2011. Participants were randomized 1:1 to either the intervention group (US plus MG) or the control group (MG alone). Randomization was conducted through individual or cluster methods depending on the site. The intervention consisted of two screening rounds over a two-year period. The primary endpoint of this specific secondary analysis was the cumulative incidence of stage 2 or higher breast cancers (TNM classification) up to the data cutoff in October 2024.

Primary Findings: Reduction in Advanced Cancers

After a median follow-up of 11.4 years in the intervention group and 11.3 years in the control group, the researchers observed a clear divergence in outcomes. In the intervention group, 894 breast cancers were detected, of which 234 (26%) were classified as advanced (Stage II+). In the control group, 843 cancers were detected, with 277 (33%) being advanced. The hazard ratio was calculated at 0.83, indicating a 17% reduction in the risk of presenting with advanced breast cancer when adjunctive US was utilized. This finding is statistically significant (p=0.026) and clinically meaningful, as Stage II+ cancers typically require more aggressive systemic therapy and have lower survival rates compared to Stage I or in-situ lesions.

Temporal Dynamics and Proportional Hazards

An intriguing aspect of the J-START data is the temporal distribution of the screening benefit. The Kaplan-Meier curves for advanced cancer incidence did not diverge immediately. Instead, they suggested a violation of the proportional hazards assumption. Significant differences in incidence were primarily localized to the period between 48 months and 96 months. The gap between the two groups began to widen around year four, reached its maximum at year eight, and remained stable thereafter. This suggests that the early detection of small, invasive cancers by US prevented these tumors from progressing to advanced stages that would have otherwise been detected clinically or by subsequent MG in the following years.

Dense Breast Tissue: The Biological Rationale

The efficacy of US in this cohort is biologically linked to the tissue characteristics of the 40–49 age group in Japan. Ultrasound utilizes high-frequency sound waves that can distinguish between solid masses (tumors) and fluid-filled cysts or dense fibroglandular tissue, which appear differently on an acoustic profile. Unlike MG, which relies on X-ray attenuation, US sensitivity is less affected by the density of the breast. By identifying tumors at an earlier, node-negative stage, US facilitates a ‘stage shift’—moving the diagnosis from a potential Stage II or III to a Stage I.

Expert Commentary

The J-START results provide the most definitive evidence to date that adjunctive US screening in women with dense breasts is not merely increasing ‘overdiagnosis’—the detection of indolent cancers that would never cause harm—but is actively reducing the incidence of advanced disease. This is a critical distinction in the debate over screening harms versus benefits.

One controversy in breast screening is whether the detection of more cancers leads to better outcomes. The reduction in Stage II+ cancers is a strong surrogate for future mortality reduction, though mortality data itself requires even longer follow-up. Some critics argue that the ‘violation of proportional hazards’ seen in the trial suggests the benefit is transient; however, the stability of the divergence after year eight argues for a lasting impact on the population’s cancer stage distribution. Clinicians should consider these findings when discussing screening options with women in their 40s, especially those of Asian descent or those known to have dense breast tissue. However, the high false-positive rate associated with US remains a challenge that requires high-quality training for sonographers and radiologists to minimize unnecessary biopsies.

Conclusion

The J-START secondary analysis confirms that adjunctive ultrasonography reduces the cumulative incidence of advanced breast cancer in women aged 40–49. With a 17% reduction in Stage II+ cases over an 11-year period, the trial underscores the potential of US to fill the diagnostic void left by mammography in dense breast tissue. These results have profound implications for global health policy, supporting the integration of US into personalized screening strategies. Future research should focus on cost-effectiveness and the further refinement of US technology, such as automated breast ultrasound (ABUS), to standardize quality across different clinical settings.

References

• Harada-Shoji N, Suzuki A, Ishida T, Yamamoto S, Kanemura S, Yamaguchi T, Shiono-Narikawa Y, Ohuchi N; J-START investigators. Cumulative incidence of advanced breast cancer in women aged 40-49 years in the Japan Strategic Anti-cancer Randomised Trial (J-START) of adjunctive ultrasonography: a prespecified secondary analysis. Lancet. 2026 Feb 21;407(10530):784-793. PMID: 41722967
• Avoidable deaths through the primary prevention, early detection, and curative treatment of cancer worldwide: a population-based study. Lancet Glob Health. 2026 Mar;14(3):e356-e366. PMID: 41713439