Oncospace

Oncospace is a software system designed to augment radiotherapy treatment planning by providing customized protocols, predicting patient-specific dosimetric goals, and automating plan optimization. It leverages updated models for various anatomical regions and natural language processing to standardize OAR names, with rigorous verification demonstrating non-inferior OAR dose sparing and maintained target coverage compared to traditional methods, ensuring safety and effectiveness while maintaining human oversight.

Oncospace, Inc. has developed a software system named Oncospace, designed to assist radiation oncologists and medical dosimetrists in radiotherapy treatment planning. This system allows for the selection and customization of treatment protocols, including dose prescriptions and dosimetric goals for both treatment targets and organs at risk (OARs). It also predicts patient-specific OAR dosimetric goals, automates the initiation of plan optimization on a treatment planning system (TPS), and provides an interface for plan evaluation. The software is intended to augment the treatment planning process, ensuring that human judgment remains central to accepting and refining goals and assessing plan quality, and it does not directly control treatment machines or make treatment decisions. The current version of Oncospace includes updated models for prostate, head, and neck regions, and has expanded its capabilities to include thoracic, abdominal, and pelvic regions for patients with malignant or benign diseases. Rigorous verification and validation testing, encompassing clinical and model performance data, was conducted. Results demonstrated that treatment plans created with Oncospace maintained target coverage and achieved non-inferior organ-at-risk dose sparing compared to traditional planning methods, without requiring additional optimization cycles. Furthermore, the system's ability to standardize organ-at-risk names using natural language processing also showed high accuracy. These tests confirm the system's safety and effectiveness.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. What specific clinical outcomes (e.g., toxicity rates, local control) have been observed in patients treated with Oncospace-generated plans compared to traditional methods in long-term follow-up studies?; How does the system's "human judgment" interface integrate into existing clinical workflows, and what training is required for oncologists and dosimetrists to effectively use and refine its outputs?; What are the limitations of the current predictive models, particularly for rare anatomies or highly complex cases, and how does Oncospace account for these uncertainties?; Can Oncospace be integrated with various treatment planning systems (TPS) from different vendors, and what is the interoperability standard used for data exchange?; Beyond dose sparing, what other quality metrics (e.g., treatment delivery time, patient setup reproducibility) are influenced by the use of Oncospace, and how are these tracked?