Ultrahypofractionated radiotherapy for localised prostate cancer: The impact of daily MRI-guided adaptive radiotherapy on delivered dose

Abstract

Introduction: Magnetic resonance image-guided adaptive radiotherapy (MRIgART) reduces uncertainties by correcting for day-to-day target and organ-at-risk deformation and motion. This is the first study to examine the dosimetric impact of MRIgART for ultrahypofractionated prostate cancer treatment, compared to standard-of-care image-guided non-adapted radiotherapy.

Methods: Twenty patients with localised prostate cancer, who received ultrahypofractionated MRIgART on the Unity MR linac (Elekta, Sweden) were retrospectively analysed. Online daily MRI was acquired for replanning (MRI_session) and a second for position verification before treatment (MRI_verification). To compare delivered dose with and without adaptation, three plans were generated offline per fraction; a session plan (reference plan adapted to MRI_session anatomy), a verification plan (session plan recalculated on MRI_verfication anatomy), and a non-adapted plan (reference plan recalculated on MRI_verfication anatomy). Target and organ-at-risk doses were calculated, and dose difference evaluated.Secondary analysis, using deformable dose accumulation, estimated verification and non-adapted dose to primary target (CTVpsv) substructures; prostate, gross tumour volume (GTV) and proximal 1 cm of seminal vesicles (1cmSV). Impact of prostate, rectum and bladder volume changes on dose were evaluated.

Results: Median dose to 95 % of the CTVpsv was significantly higher with adaptation; 40.3, 40.0 and 38.2 Gy for session, verification, and non-adapted plans. Adaptation achieved a lower median urethra V42Gy dose but bladder V37Gy dose was lower when not adapting. Rectum V36Gy dose was similar for adapted and non-adapted plans.CTVpsv substructure dose difference was greatest for 1cmSV; 40.0 versus 37.5 Gy for verification/non-adapted plans. Adaptation achieved significantly higher prostate only, but not GTV doses. Prostate and rectal volume changes had a negative impact on non-adapted dose only.

Conclusion: MRIgART, offers significant dosimetric benefit for ultrahypofractionated prostate cancer compared to non-adapted strategies. Greatest benefit is expected for those with SV or high-risk of SV involvement, persistent rectal gas, prostate swelling and for the application of novel dose strategies including GTV dose escalation and non-involved prostate dose de-escalation.

Keywords: Adaptive radiotherapy; MRI-guided radiotherapy; Prostate cancer; Ultrahypofractionated radiotherapy.

Fig. 1

Overview of plan generation process.

Fig. 2

A (top): Median, interquartile and range dose to key target and OAR constraints. ns = not significant, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001. Green dotted line = optimal dose constraint, purple dotted line = mandatory dose constraint. Fig. 2B (bottom): Median dose to CTVpsv_40. Error bars represent the minimum and maximum daily plan doses, as though each plan delivered the entire treatment course. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

A (top): Median, interquartile and range dose to key target and OAR constraints. ns = not significant, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001. Green dotted line = optimal dose constraint, purple dotted line = mandatory dose constraint. Fig. 2B (bottom): Median dose to CTVpsv_40. Error bars represent the minimum and maximum daily plan doses, as though each plan delivered the entire treatment course. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Group estimated cumulative dose to the CTVpsv and its substructures. Orange plots = verification plans, grey plots = non-adapted plans. ns = not significant, **** = p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

The only high and moderate Spearman’s correlation results were found for non-adapted plan CTVpsv substructure doses and prostate and rectum volume changes. rs ≥ 0.5 = moderate correlation, rs ≥ 0.7 = high correlation.

Fig. 5

A shows patient 20 and 5, the non-adapted plan for #4 is on the left and the verification plan for #4 the right, red isodose = 40 Gy, orange isodose = 36.25 Gy. Cumulative prostate (yellow contour) and GTV (blue contour) dose dropped below 36.25 Gy on the non-adapted plan for both. B a sagittal view of the planning scan for patients 1, 3, 6 and 15 (top) and from #5 of their non-adapted plans (bottom), the white arrow indicates missed SVs (green contour) in the absence of adaptation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)