Patient-specific quality assurance (PSQA) is an important piece of the radiotherapy clinical workflow. Per the Green Journal, PSQA refers to the systematic verification that a patient’s treatment plan can be safely and accurately delivered.

Today, there are important questions about PSQA, including how often it detects failures, how to integrate it into everyday workflows without reducing a clinic’s capacity to treat patients, and incorporating more on-treatment and in vivo QA procedures for a more comprehensive patient-specific validation.  

Approaches for PSQA

There are several conventional approaches to PSQA, with benefits and drawbacks to each. These include:

• Phantom-Based – PSQA with phantoms involves delivering the clinical plan to a phantom, a physical measurement device with a plastic shell or composite material with a measurement tool: film, ion chambers, diodes, or an array. Phantoms are time-consuming for PSQA, as they require use of the delivery linac, reducing the number of patients a clinic can process or necessitating after-hours work.
• Portal Dosimetry – Imager-based PSQA uses electronic portal images to get information on the actual dose delivered to the patient. This can also give unique insight into real-time dose delivery for that patient, which is of particular interest as adaptive workflows come online. However, EPID-based in vivo images have some practical limitations, including poor detector quality, subpar dose reconstruction algorithms, and limited commercial solutions.
• Log File Analysis – Log files collected from treatment delivery systems are an attractive alternative to phantoms. As log files are automatically collected from the machine, it is relatively low effort to use them to analyze delivered dose and conduct PSQA without using valuable machine time or manually reviewing images, and can be used in both pre-treatment and on-treatment settings.
• Artificial Intelligence – With the growing applications of AI in the radiotherapy clinic and the large volume of historical data available for analysis, there is increasing attention toward the development of AI models for conducting PSQA – before the dose is even delivered. Methods for this are still nascent, and there are concerns about edge-case failures missed due to lack of models, but it remains an area of research interest, especially to accelerate online adaptive workflows.

Investigating failure modalities that can be prevented with PSQA

Typically, research recommends dose delivery uncertainty be kept below 5%, however uncontrollable factors likely add to error rates. Recently, AAPM TG-360 undertook an international, multi-clinic, multi-TPS investigation to determine how IMRT and VMAT treatments are impacted by different modes of failure.

The group based its evaluations on the TG-100 report and weighted each failure mode based on severity, occurrence, and detectability. Based on this, they assigned each error an “RPN” score, or risk priority number. Higher numbers are associated with increased risk.

The highest RPN issues for PSQA were found to be slightly lower on the severity ranking, but they occurred frequently and were difficult to catch.

The top three RPN failures were:

• Output factor incorrect: RPN 315 (MLC) or 252 (jaws) - Result: wrong dose distribution
• MLC modeling error: RPN 315 - Result: wrong absolute dose
• PDD/TMR incorrect: RPN 252 - Result: wrong dose distribution

TG-360 demonstrated that there are a number of potential errors when delivering treatments. The group noted that PSQA is often overlooked in the overall radiation therapy process, so it does not always receive high consideration in clinics. However, the work shows that PSQA is still a critical part of the treatment process and, when properly implemented, can reduce risks to patients.

How Radformation equips clinics for PSQA

ClearCalc, with an optional RadMonteCarlo add-on, supports log file analysis for PSQA.

Launch your log file analysis results directly from the standalone dashboard with the Zero-Click dashboard.

Review your results at a glance in the workspace.

Validate your plans against finite-sized pencil beam or Monte Carlo algorithm* results based on your delivery log files.

Leverage the user-friendly interface to review plan profiles.

Don’t discount the importance of PSQA for patient safety

While there are a number of ways to approach PSQA in the modern radiotherapy clinic, log file analysis offers a notably fast and comprehensive solution. With log files, clinicians can recalculate delivered dose and even implement the highest accuracy for recalculations with Monte Carlo algorithms. The result is a PSQA program that ensures a compliant clinic without slowing treatment delivery.

PSQA fits into a broader practice of quality assurance including machine QA, regular audits, and clear interdepartmental communication, which also support patient safety and should not be overlooked.

To learn more about what ClearCalc with the optional RadMonteCarlo add-on can do to transform your PSQA workflows, enhance clinical safety, and save valuable time in your department, contact our experts to schedule a personalized demo today.

*Monte Carlo algorithm is available as part of RadMonteCarlo, an optional add-on for ClearCalc.