Scholarlink

Objectives <p>To evaluate whether pretreatment magnetic resonance imaging (MRI) parameters can predict short-term clinical response to dienogest (DNG) in patients with adenomyosis.</p> Methods <p>This retrospective study included 78 patients with MRI-diagnosed adenomyosis who underwent pelvic MRI before hormonal therapy between October 2018 and July 2025. Quantitative MRI parameters included T2 signal intensity ratios, diffusion-weighted imaging (DWI) signal intensity ratios, normalized apparent diffusion coefficient (ADC), and uterine morphological parameters. Adenomyosis subtypes were classified according to the modified Kishi criteria. Short-term clinical response was assessed primarily 3–6&#xa0;months after treatment initiation as a composite clinical outcome encompassing symptom improvement (dysmenorrhea, menstrual blood loss, and/or hemoglobin levels) and treatment continuation; patients were classified as responders or non-responders. Predictive analyses were restricted to the DNG cohort because only one patient in the GnRH cohort was a non-responder.</p> Results <p>Of the 78 patients, 32 received gonadotropin-releasing hormone (GnRH) agonist or antagonist therapy and 46 received DNG. In the DNG cohort, 30 patients were responders and 16 were non-responders. MRI-based adenomyosis subtype, lesion distribution, and uterine morphological parameters were not significantly associated with response in DNG-treated patients. However, absolute ADC values were significantly higher in responders (1.03 vs. 0.89 × 10⁻<sup>3</sup>&#xa0;mm<sup>2</sup>/s, P = 0.036), as was the ADC signal intensity ratio relative to the endometrium (ADC SIR<sub>endo</sub>: 0.92 vs. 0.85, P = 0.034). Receiver operating characteristic analysis demonstrated moderate discrimination for both parameters (area under the curve = 0.70). Optimal cut-off values were 0.951 × 10⁻<sup>3</sup>&#xa0;mm<sup>2</sup>/s for ADC and 0.952 for ADC SIR<sub>endo</sub>.</p> Conclusion <p>Quantitative diffusion MRI parameters were associated with short-term clinical response to DNG, whereas conventional morphological features were not. Diffusion-weighted MRI may provide complementary imaging biomarkers for adenomyosis stratification in DNG-treated patients; prospective validation with longer follow-up is warranted.</p> Graphical abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Pretreatment MRI parameters as predictive biomarkers for short-term clinical response to dienogest in adenomyosis: a retrospective cohort study

  • 期刊论文

  • 发表日期：

    2026年6月18日

查看全文
  • Kazuhiko Morikawa,
  • Akira Baba,
  • Shun Kusada,
  • Satoshi Matsushima,
  • Yohei Ohki,
  • Megumi Shiraishi,
  • Yoshitake Miyamoto,
  • Aya Igarashi,
  • Yumari Kusano,
  • Yuki Hayakawa,
  • Ayako Kawabata,
  • Hiroya Ojiri

摘要

Objectives

To evaluate whether pretreatment magnetic resonance imaging (MRI) parameters can predict short-term clinical response to dienogest (DNG) in patients with adenomyosis.

Methods

This retrospective study included 78 patients with MRI-diagnosed adenomyosis who underwent pelvic MRI before hormonal therapy between October 2018 and July 2025. Quantitative MRI parameters included T2 signal intensity ratios, diffusion-weighted imaging (DWI) signal intensity ratios, normalized apparent diffusion coefficient (ADC), and uterine morphological parameters. Adenomyosis subtypes were classified according to the modified Kishi criteria. Short-term clinical response was assessed primarily 3–6 months after treatment initiation as a composite clinical outcome encompassing symptom improvement (dysmenorrhea, menstrual blood loss, and/or hemoglobin levels) and treatment continuation; patients were classified as responders or non-responders. Predictive analyses were restricted to the DNG cohort because only one patient in the GnRH cohort was a non-responder.

Results

Of the 78 patients, 32 received gonadotropin-releasing hormone (GnRH) agonist or antagonist therapy and 46 received DNG. In the DNG cohort, 30 patients were responders and 16 were non-responders. MRI-based adenomyosis subtype, lesion distribution, and uterine morphological parameters were not significantly associated with response in DNG-treated patients. However, absolute ADC values were significantly higher in responders (1.03 vs. 0.89 × 10⁻3 mm2/s, P = 0.036), as was the ADC signal intensity ratio relative to the endometrium (ADC SIRendo: 0.92 vs. 0.85, P = 0.034). Receiver operating characteristic analysis demonstrated moderate discrimination for both parameters (area under the curve = 0.70). Optimal cut-off values were 0.951 × 10⁻3 mm2/s for ADC and 0.952 for ADC SIRendo.

Conclusion

Quantitative diffusion MRI parameters were associated with short-term clinical response to DNG, whereas conventional morphological features were not. Diffusion-weighted MRI may provide complementary imaging biomarkers for adenomyosis stratification in DNG-treated patients; prospective validation with longer follow-up is warranted.

Graphical abstract