Breath-hold protons lower heart dose but increase lung exposure in breast cancer

In 15 left-sided breast cancer cases, proton breath-hold reduced median heart dose from 2.1 to 0.8 Gy but increased lung exposure.

KEY POINTS

  • This single-center in-silico planning study from Maastro, the Netherlands, evaluated 15 patients with left-sided breast cancer requiring internal mammary nodal irradiation, selected from a larger cohort of 234 patients considered for proton therapy between March 2019 and October 2024. Each patient had planning CT acquired during free breathing (FB) and deep inspiration breath-hold (BH). Three plans were compared: the clinically generated photon BH plan, a retrospectively generated IMPT FB plan, and an IMPT BH plan. Mean patient age was 51.3 years, and 12/15 (80%) had cardiovascular risk factors.
  • Proton plans maintained equivalent target coverage while incorporating robust optimization. Treatment volumes included the breast or chest wall, internal mammary nodes, and other indicated regional lymphatics. Eleven patients (73.3%) had a prescription of 40.05 Gy in 15 fractions, while four received 20-fraction treatment with a simultaneous integrated tumor-bed boost. IMPT used two or three beam angles, with 5-mm setup and 3% range uncertainty across 28 robustness scenarios. All proton plans satisfied institutional robust CTV coverage criteria, including D98% ≥93% for elective CTVs and ≥94% for boost CTVs.
  • Breath-hold substantially reduced mean heart dose even when proton therapy was already used. Median mean heart dose was 5.0 Gy (IQR 3.9–6.0) for photon BH, 2.1 Gy (IQR 1.3–3.7) for proton FB, and 0.8 Gy (IQR 0.6–1.8) for proton BH. The additional median reduction from proton FB to BH was approximately 1.3 Gy, and differences between the planning approaches were statistically significant (p<0.01). Mean contralateral breast dose was also lower with protons, measuring 0.2 Gy with both proton approaches versus 1.7 Gy with photon BH.
  • The cardiac advantage was accompanied by higher mean lung dose in most patients. Median mean lung dose increased from 5.4 Gy (IQR 4.1–6.6) with proton FB to 6.7 Gy (IQR 5.4–7.5) with proton BH (p<0.05). Fourteen of 15 patients had higher lung dose during breath-hold, with a median individual increase of 0.85 Gy (range −0.16 to +3.37 Gy). In four patients, proton BH plans delivered a higher mean lung dose than the corresponding photon BH plans, demonstrating that breath-hold did not uniformly improve all OAR metrics.
  • Modeled cardiac benefit was greatest in younger patients with cardiovascular risk factors. Using the Dutch national model-based proton selection framework, investigators simulated female patients aged 40–80 years, with and without cardiac risk factors, using the 15 paired treatment anatomies. Proton BH consistently reduced predicted lifetime acute coronary event risk compared with proton FB (all p≤0.002). The subgroup aged ≤60 years with cardiac risk factors was identified as most likely to achieve a clinically meaningful additional absolute risk reduction of approximately 1 percentage point or more, whereas modeled benefits were generally smaller in older patients and those without cardiovascular risk factors.
  • Higher lung exposure could offset the cardiac benefit in younger smokers. For simulated smokers aged ≤50 years, the median modeled reduction in secondary lung cancer risk compared with photon BH was 1.5 percentage points with proton FB, but only 0.3 percentage points with proton BH. When predicted cardiac and secondary lung cancer benefits were considered together, the median overall advantage of proton BH over proton FB approached zero in these modeled comparisons. Nevertheless, under the Dutch ≥2-percentage-point NTCP reduction threshold for proton eligibility, 6/15 patients qualified using proton FB, while 13/15 qualified using proton BH, representing seven additional potentially eligible patients.
  • The findings remain planning- and model-based rather than clinically demonstrated. Breath-hold increased average lung volume by approximately 1.4 liters and reduced lung CT density by a median 123 HU, potentially contributing to changes in proton range and lung exposure. Implementing supported proton BH also requires additional imaging, surface monitoring, and potentially nasal high-flow therapy; the authors' prior lymphoma experience suggested approximately 10 additional minutes per fraction, but this was not measured in the breast cohort. The small sample, arbitrarily selected 1% incremental NTCP threshold, proton NTCP models lacking direct validation, and reliance on one proton delivery system limit generalizability.

CLINICAL TAKEAWAY

Deep inspiration breath-hold may provide additional cardiac sparing during IMPT for selected patients with left-sided breast cancer requiring internal mammary nodal irradiation, particularly younger patients with cardiovascular risk factors. However, the associated increase in lung dose and modeled secondary lung cancer risk in younger smokers support individualized FB-versus-BH proton plan comparisons rather than routine breath-hold use for all proton-treated patients.

SOURCE

Technical Innovations & Patient Support in Radiation Oncology

Browse more research Suggest a correction