Dataset Information

Highly Efficient Training, Refinement, and Validation of a Knowledge-based Planning Quality-Control System for Radiation Therapy Clinical Trials.

ABSTRACT: To demonstrate an efficient method for training and validation of a knowledge-based planning (KBP) system as a radiation therapy clinical trial plan quality-control system.We analyzed 86 patients with stage IB through IVA cervical cancer treated with intensity modulated radiation therapy at 2 institutions according to the standards of the INTERTECC (International Evaluation of Radiotherapy Technology Effectiveness in Cervical Cancer, National Clinical Trials Network identifier: 01554397) protocol. The protocol used a planning target volume and 2 primary organs at risk: pelvic bone marrow (PBM) and bowel. Secondary organs at risk were rectum and bladder. Initial unfiltered dose-volume histogram (DVH) estimation models were trained using all 86 plans. Refined training sets were created by removing sub-optimal plans from the unfiltered sample, and DVH estimation models… and DVH estimation models were constructed by identifying 30 of 86 plans emphasizing PBM sparing (comparing protocol-specified dosimetric cutpoints V10 (percentage volume of PBM receiving at least 10 Gy dose) and V20 (percentage volume of PBM receiving at least 20 Gy dose) with unfiltered predictions) and another 30 of 86 plans emphasizing bowel sparing (comparing V40 (absolute volume of bowel receiving at least 40 Gy dose) and V45 (absolute volume of bowel receiving at least 45 Gy dose), 9 in common with the PBM set). To obtain deliverable KBP plans, refined models must inform patient-specific optimization objectives and/or priorities (an auto-planning "routine"). Four candidate routines emphasizing different tradeoffs were composed, and a script was developed to automatically re-plan multiple patients with each routine. After selection of the routine that best met protocol objectives in the 51-patient training sample (KBPFINAL), protocol-specific DVH metrics and normal tissue complication probability were compared for original versus KBPFINAL plans across the 35-patient validation set. Paired t tests were used to test differences between planning sets.KBPFINAL plans outperformed manual planning across the validation set in all protocol-specific DVH cutpoints. The mean normal tissue complication probability for gastrointestinal toxicity was lower for KBPFINAL versus validation-set plans (48.7% vs 53.8%, P<.001). Similarly, the estimated mean white blood cell count nadir was higher (2.77 vs 2.49 k/mL, P<.001) with KBPFINAL plans, indicating lowered probability of hematologic toxicity.This work demonstrates that a KBP system can be efficiently trained and refined for use in radiation therapy clinical trials with minimal effort. This patient-specific plan quality control resulted in improvements on protocol-specific dosimetric endpoints.

SUBMITTER: Li N

PROVIDER: S-EPMC5175211 | biostudies-literature | 2017 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Highly Efficient Training, Refinement, and Validation of a Knowledge-based Planning Quality-Control System for Radiation Therapy Clinical Trials.

Li Nan N Carmona Ruben R Sirak Igor I Kasaova Linda L Followill David D Michalski Jeff J Bosch Walter W Straube William W Mell Loren K LK Moore Kevin L KL

International journal of radiation oncology, biology, physics 20161013 1

<h4>Purpose</h4>To demonstrate an efficient method for training and validation of a knowledge-based planning (KBP) system as a radiation therapy clinical trial plan quality-control system.<h4>Methods and materials</h4>We analyzed 86 patients with stage IB through IVA cervical cancer treated with intensity modulated radiation therapy at 2 institutions according to the standards of the INTERTECC (International Evaluation of Radiotherapy Technology Effectiveness in Cervical Cancer, National Clinica ...[more]

PMID: 27979445

Shared Molecules

Only show the datasets with similarity scores above: 0.5

Threshold

0.5

Similar Datasets

Project description:PURPOSE:This study aimed to evaluate the feasibility of using a single-institution, knowledge-based planning (KBP) model as a dosimetric plan quality control (QC) for multi-institutional clinical trials. The efficacy of this QC tool was retrospectively evaluated using a subset of plans submitted to Radiation Therapy Oncology Group (RTOG) study 0617. METHODS AND MATERIALS:A single KBP model was created using commercially available software (RapidPlan; Varian Medical Systems, Palo Alto, CA) and data from 106 patients with non-small cell lung cancer who were treated at a single institution. All plans had prescriptions that ranged from 60 Gy in 30 fractions to 74 Gy in 37 fractions and followed the planning guidelines from RTOG 0617. Two sets of optimization objectives were created to produce different trade-offs using the single KBP model predictions: one prioritizing target coverage and a second prioritizing lung sparing (LS) while allowing an acceptable variation in target coverage. Three institutions submitted a high volume of clinical plans to RTOG 0617 and provided data on 25 patients, which were replanned using both sets of optimization objectives. Model-generated, dose-volume histogram predictions were used to identify patients who exceeded the lung clinical target volume (CTV) V20Gy >37% and would benefit from the LS objectives. Overall plan quality differences between KBP-generated plans and clinical plans were evaluated at RTOG 0617-defined dosimetric endpoints. RESULTS:Target coverage and organ at risk sparing was significantly improved for most KBP-generated plans compared with those from clinical trial data. The KBP model using prioritized target coverage objectives reduced heart Dmean and V40Gy by 2.1 Gy and 5.2%, respectively. Similarly, using LS objectives reduced the lung CTV Dmean and V20Gy by 2.0 Gy and 2.9%, respectively. The KBP predictions correctly identified all patients with lung CTV V20Gy > 37% (5 of 25 patients) and significantly reduced the dose to the lung CTV by applying the LS optimization objectives. CONCLUSIONS:A single-institution KBP model can be applied as a QC tool for multi-institutional clinical trials to improve overall plan quality and provide decision-support to determine the need for anatomy-based dosimetric trade-offs.

Dataset Information

Highly Efficient Training, Refinement, and Validation of a Knowledge-based Planning Quality-Control System for Radiation Therapy Clinical Trials.

Publications

Highly Efficient Training, Refinement, and Validation of a Knowledge-based Planning Quality-Control System for Radiation Therapy Clinical Trials.

Shared Molecules

Only show the datasets with similarity scores above: 0.5

Threshold

0.5

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets