Evaluation of a Novel Quantitative Multiparametric MR Sequence for Radiation Therapy Treatment Response Assessment.

Abstract

Background: Multi-parametric MRI has shown great promise to rapidly derive multiple quantitative imaging biomarkers for treatment response assessment.

Purpose: To evaluate a novel Deep-Learning-enhanced MUlti-PArametric MR sequence (DL-MUPA) for treatment response assessment for brain metastases patients treated with stereotactic radiosurgery (SRS) and head-and-neck (HnN) cancer patients undergoing conventionally fractionation adaptive radiation therapy.

Methods: DL-MUPA derives quantitative T1 and T2 relaxation time maps from a single 4-6-minute scan denoised via DL method using least-squares dictionary fitting. Longitudinal phantom benchmarking was performed on a NIST-ISMRM phantom over one year. In patients, longitudinal DL-MUPA data were acquired on a 1.5T MR-simulator, including pre-treatment (PreTx) and every ~3 months after SRS (PostTx) in brain, and PreTx, mid-treatment and 3 months PostTx in HnN. Delta analysis was performed calculating changes of mean T1 and T2 values within gross tumor volumes (GTVs), residual disease (RD, HnN), parotids, and submandibular glands (HnN) for treatment response assessment. Uninvolved normal tissues (normal appearing white matter in brain, masseter in HnN) were evaluated to quantify within-subject repeatability.

Results: Phantom benchmarking revealed excellent inter-session repeatability (coefficient of variance <0.9% for T1, <6.6% for T2), suggesting reliability for longitudinal studies once systematic biases are adjusted. Uninvolved normal tissue suggested acceptable within-subject repeatability (brain |ΔT1_mean|<36ms/5.0%, |ΔT2_mean|<2ms/5.0%, HnN |ΔT1_mean|<69ms/7.0%, |ΔT2_mean|<4ms/17.8% due to low T2). In brain, remarkable changes were noted in resolved metastasis (4-month PostTx ΔT1_mean=155ms/13.7%) and necrotic settings (ΔT1_mean=214-502ms/17.6-39.7%, ΔT2_mean=7-41ms/8.7-41.4%, 6-month to 3-month PostTx). In HnN, two base of tongue tumors exhibited T2 enhancement (PostTx GTV ΔT2_mean>7ms/12.8%, RD ΔT2_mean>10ms/18.1%). A case with nodal disease resolved PostTx (GTV ΔT1_mean=-541ms/-39.5%, ΔT2_mean=-24ms/-32.7%, RD ΔT1_mean=-400ms/-29.2%, ΔT2_mean=-25ms/-35.3%). Enhancement was found in involved parotids (PostTx ΔT1_mean>82ms/12.4%, ΔT2_mean>6ms/13.4%) and submandibular glands (PostTx ΔT1_mean>135ms/14.6%, ΔT2_mean>17ms/34.5%) while the uninvolved organs remained stable.

Conclusions: Preliminary results suggest promise of DL-MUPA for treatment response assessment and highlight potential endpoints for functional sparing.