Radiation Planning Assistant (RPA)

K222728 · University of Texas, MD Anderson Cancer Center · MUJ · May 17, 2023 · Radiology

Device Facts

• Record ID: K222728
• Device Name: Radiation Planning Assistant (RPA)
• Applicant: University of Texas, MD Anderson Cancer Center
• Product Code: MUJ · Radiology
• Decision Date: May 17, 2023
• Decision: SESE
• Submission Type: Traditional
• Regulation: 21 CFR 892.5050
• Device Class: Class 2
• Attributes: AI/ML, Software as a Medical Device

AI Performance

Output	Acceptance	Observed	Dev DS	Test DS
Cervix Radiotherapy Treatment Planning (Normal Structures)	Difference in passing rates between RPA plan and clinical plan ≤ 5%.	< 5% difference for all bony structures and critical soft tissue structures (with VMAT and 4-field box).	1,999 patients (2,254 CT scans) for normal tissue; 192 patients (316 CT scans) for secondary normal tissue; 406-490 CT scans for various CTVs; 119 patients (169 CT scans) for liver.	50 patients (VMAT) and 47/45 patients (3D) across 5 clinical sites; datasets collected starting Jan 1, 2022 (or Jan 1, 2021 if insufficient).
Cervix Radiotherapy Treatment Planning (Target Structures)	Difference in passing rates between RPA plan and clinical plan ≤ 5%.	< 5% difference for all assessed structures.	1,999 patients (2,254 CT scans) for normal tissue; 192 patients (316 CT scans) for secondary normal tissue; 406-490 CT scans for various CTVs; 119 patients (169 CT scans) for liver.	50 patients (VMAT) and 47/45 patients (3D) across 5 clinical sites; datasets collected starting Jan 1, 2022 (or Jan 1, 2021 if insufficient).
Chest Wall Radiotherapy Treatment Planning	Difference in passing rates between RPA plan and clinical plan ≤ 5% (effectiveness) and ≤ 7% (safety).	≤ 5% difference (effectiveness) and ≤ 7% difference (safety) for all assessed structures.	250 patients treated between August 2016 and June 2021.	46 patients across 5 clinical sites; datasets collected starting Jan 1, 2022 (or Jan 1, 2021 if insufficient).
Head and Neck Radiotherapy Treatment Planning	Difference in passing rates between RPA plan and clinical plan ≤ 5%.	< 5% difference for the majority of assessed structures.	3,288 patients (3,495 CT scans) for primary normal tissue; 160 patients for secondary normal tissue; 61 patients for lymph node CTVs.	86 patients across 5 clinical sites; datasets collected starting Jan 1, 2022 (or Jan 1, 2021 if insufficient).
Whole Brain Radiotherapy Treatment Planning	Difference in passing rates between RPA plan and clinical plan ≤ 5%.	< 6% difference (safety) and < 9% difference (effectiveness) for lenses; < 5% difference for target structures.	1,966 spinal canal CNN; 803 VB labeling; 107 VB segmentation from 930 MDACC and 355 external patients.	46 patients across 5 clinical sites; datasets collected starting Jan 1, 2022 (or Jan 1, 2021 if insufficient).

Indications for Use

The Radiation Planning Assistant (RPA) is used to plan radiotherapy treatments for patients with cancers of the head and neck, cervix, breast, and metastases to the brain. The RPA is used to plan external beam irradiation with photon beams using computerized tomography (CT) images. The RPA is used to create contours and treatment plans that the user imports into their own Treatment Planning System (TPS) for review, editing, and re-calculation of the dose. Some functions of the RPA use Eclipse 15.6. The RPA is not intended to be used as a primary treatment planning system. All automatically generated contours and plans must be imported into the user's own treatment planning system for review, edit, and final dose calculation.

Device Story

RPA is a web-based radiotherapy planning tool; inputs CT images; utilizes deep learning and automated algorithms to generate contours and treatment plans; integrates with Eclipse 15.6 for specific functions; output consists of contours and plans for import into a primary Treatment Planning System (TPS). Used by clinicians in oncology settings; output requires review, editing, and final dose calculation by the user in their own TPS. Benefits include automated contouring and planning efficiency; assists in radiotherapy workflow; not a primary TPS.

Clinical Evidence

Retrospective multicenter study; 5 clinical sites; evaluated 50 cervix VMAT, 47 cervix 3D, 46 chest wall, 86 head and neck, and 46 whole brain patient plans. Primary endpoints: dosimetric metric passing rates (RPA vs. clinical plans) and geometric accuracy (recall, DSC, isocenter agreement). Results showed <5-9% difference in dosimetric passing rates for most structures; 25th percentile recall >0.7; Surface DSC >0.8 for 95% of scans; isocenter agreement <3mm. No clinical prospective data.

Technological Characteristics

Web-based software; utilizes deep learning algorithms for contouring and planning; integrates with Eclipse 15.6 (K181145). Conforms to IEC 62304 (software lifecycle) and IEC 62083 (radiotherapy safety). Operates on CT image inputs; provides automated contouring, dose optimization, and quality control checks.

Indications for Use

Indicated for patients with cancer of the head and neck, cervix, breast, and metastases to the brain requiring external beam photon radiotherapy planning.

Regulatory Classification

Identification

A medical charged-particle radiation therapy system is a device that produces by acceleration high energy charged particles (e.g., electrons and protons) intended for use in radiation therapy. This generic type of device may include signal analysis and display equipment, patient and equipment supports, treatment planning computer programs, component parts, and accessories.

Predicate Devices

• Eclipse Treatment Planning System v15.6 (K181145)

Submission Summary (Full Text)

{0}------------------------------------------------

May 17, 2023

Image /page/0/Picture/1 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). On the left is the Department of Health & Human Services logo. To the right of that is the FDA logo, which is a blue square with the letters "FDA" in white. To the right of the blue square is the text "U.S. FOOD & DRUG ADMINISTRATION" in blue.

University of Texas, MD Anderson Cancer Center % Ms. Stella Tsai Sr. Project Manager 1515 Holcombe Blvd. HOUSTON TX 77030

Re: K222728

Trade/Device Name: Radiation Planning Assistant (RPA) Regulation Number: 21 CFR 892.5050 Regulation Name: Medical charged-particle radiation therapy system Regulatory Class: Class II Product Code: MUJ Dated: April 17, 2023 Received: April 17, 2023

Dear Ms. Stella Tsai:

We have reviewed your Section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database located at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.

If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.

Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR 803) for devices or postmarketing safety reporting (21 CFR 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reporting

{1}------------------------------------------------

combination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.

Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.

For comprehensive regulatory information about medical devices and radiation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email ([e-mail]) or phone [phone] or 301-796-7100).

Sincerely,

Image /page/1/Picture/5 description: The image shows a digital signature. The signature is from Lora D. Weidner -S. The date of the signature is 2023.05.17 and the time is 07:14:45-04'00'.

Lora D. Weidner, Ph.D. Assistant Director Radiation Therapy Team DHT8C: Division of Radiological Imaging and Radiation Therapy Devices OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

{2}------------------------------------------------

# Indications for Use

510(k) Number (if known) K222728

Device Name Radiation Planning Assistant (RPA)

### Indications for Use (Describe)

The Radiation Planning Assistant (RPA) is used to plan radiotherapy treatments with cancers of the head and neck, cervix, breast, and metastases to the brain. The RPA is used to plan external beam irradiation with photon beams using CT images. The RPA is used to create contours and treatment plans that the user imports into their own Treatment Planning System (TPS) for review, editing, and re-calculation of the dose.

Some functions of the RPA use Eclipse 15.6. The RPA is not intended to be used as a primary treatment planning system. All automatically generated contours and plans must be imported into the user's own treatment planning system for review, edit, and final dose calculation.

| Type of Use (Select one or both, as applicable) | |
|-------------------------------------------------|--|
|-------------------------------------------------|--|

X Prescription Use (Part 21 CFR 801 Subpart D)

Over-The-Counter Use (21 CFR 801 Subpart C)

# CONTINUE ON A SEPARATE PAGE IF NEEDED.

This section applies only to requirements of the Paperwork Reduction Act of 1995.

# *DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.*

The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:

> Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov

"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."

{3}------------------------------------------------

# K222728

# DATE PREPARED: 16 May 2023

#### SUBMITTER 1.

| Manufacturer Name: | The University of Texas MD Anderson Cancer Center<br>Department of Radiation Physics<br>Division of Radiation Oncology<br>1515 Holcombe Blvd.<br>Houston, TX 77030 |
|--------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Official Contact: | Stella Tsai, MHA, CCRA<br>Sr. Project Director, IND Office<br>The University of Texas MD Anderson Cancer Center<br>1515 Holcombe Blvd. Unit 1634<br>Houston, TX 77030<br>Telephone [phone]<br>[e-mail] |

### 2. DEVICE

| Name of Device: | Radiation Planning Assistant (RPA) |
|-----------------------|-----------------------------------------------------------------------|
| Common or Usual Name: | System, Planning, Radiation Therapy Treatment |
| Classification Name: | 21CFR 892.5050 - Medical charged-particle radiation<br>therapy system |
| Regulatory Class: | II |
| Product Code: | MUJ |

#### PREDICATE DEVICE 3.

Eclipse Treatment Planning System v15.6 (K181145)

#### 4. DEVICE DESCRIPTION

# Design Characteristics

The Radiation Planning Assistant (RPA) is a web-based contouring and radiotherapy treatment planning software tool that incorporates the basic radiation planning functions from automated contouring, automated planning with dose optimization, and quality control checks. The system is intended for use for patients with cancer of the head and neck, cervix, breast, and metastases to the brain. The RPA system is integrated with the Eclipse Treatment Planning System v15.6 software cleared under K181145. The RPA radiation treatment planning software tool was trained against hundreds / thousands of CT Scans of normal and diseased tissues from patients receiving radiation for head and neck, cervical, breast, and whole brain at MD Anderson Cancer Center.

{4}------------------------------------------------

### 5. INDICATIONS FOR USE

The Radiation Planning Assistant (RPA) is used to plan radiotherapy treatments for patients with cancers of the head and neck, cervix, breast, and metastases to the brain.

The RPA is used to plan external beam irradiation with photon beams using computerized tomography (CT) images. The RPA is used to create contours and treatment plans that the user imports into their own Treatment Planning System (TPS) for review, editing, and re-calculation of the dose.

Some functions of the RPA use Eclipse 15.6. The RPA is not intended to be used as a primary treatment planning system. All automatically generated contours and plans must be imported into the user's own treatment planning system for review, edit, and final dose calculation.

### COMPARISON OF TECHNOLOGICAL CHARACTERISTICS 6. WITH THE PREDICATE DEVICE

The Radiation Planning Assistant (RPA) is substantially equivalent to the Eclipse Treatment Planning System v15.6 (K181145) predicate device in the following respects:

{5}------------------------------------------------

| | Subject Device | Predicate Device |
|------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| | Radiation Planning Assistant (RPA) | Eclipse Treatment Planning System Version<br>15.6 |
| | | K181145 |
| CFR<br>Citation | 892.5050 | 892.5050 |
| Product<br>Code | MUJ | MUJ |
| Indications<br>for Use | The Radiation Planning Assistant (RPA) is<br>used to plan radiotherapy treatments for<br>patients with cancers of the head and neck,<br>cervix, breast, and metastases to the brain. The<br>RPA is used to plan external beam irradiation<br>with photon beams using computerized<br>tomography (CT) images. The RPA is used to<br>create contours and treatment plans that the<br>user imports into their own Treatment Planning<br>System (TPS) for review, editing, and re-<br>calculation of the dose.<br>Some functions of the RPA use Eclipse<br>v.15.6. The RPA is not intended to be used as | The Eclipse Treatment Planning System<br>(Eclipse TPS) is used to plan radiotherapy<br>treatments for patients with malignant or<br>benign diseases. Eclipse TPS is used to plan<br>external beam irradiation with photon, electron,<br>and proton beams, as well as for internal<br>irradiation (brachytherapy) treatments. |
| | a primary treatment planning system. All<br>automatically generated contours and plans<br>must be imported into the user's own<br>treatment planning system for review, edit,<br>and final dose calculation. | |
| Device<br>Description | The Radiation Planning Assistant (RPA) is a<br>web-based contouring and radiotherapy<br>treatment planning software tool that<br>incorporates the basic radiation planning<br>functions from automated contouring,<br>automated planning with dose optimization,<br>and quality control checks. The system is<br>intended for use for patients with cancer of the<br>head and neck, cervix, breast, and metastases to<br>the brain. The RPA system is integrated with<br>the Eclipse Treatment Planning System v.15.6<br>software cleared under K181145. | The Varian Eclipse™ Treatment Planning<br>System (Eclipse TPS) provides software tools<br>for planning the treatment of malignant or<br>benign diseases with radiation. Eclipse TPS is<br>a computer-based software device used by<br>trained medical professionals to design and<br>simulate radiation therapy treatments.<br>Eclipse TPS is capable of planning treatments<br>for external beam irradiation with photon,<br>electron, and proton beams, as well as for<br>internal irradiation (brachytherapy) treatments. |

Table 1: Comparison of the Technological Characteristics of the RPA with Predicate Device

{6}------------------------------------------------

| Software<br>Function | Description of Functions Available in<br>Eclipse | Differences | Similarities | Rationale for<br>Substantial Equivalence |
|------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Software<br>Contouring<br>Functions | 1. Organ-specific autocontouring algorithms. Eclipse<br>v.15.6 includes the following organ-specific<br>autocontouring algorithms for the following organs:<br>spine, lung, brain, eye, bone.<br>2. Expert Segmentation. Eclipse v.15.6 includes an atlas-<br>based contouring approach ("expert segmentation") for<br>autocontouring of many structures, including:<br>Head / neck region: Body, bones, brainstem,<br>cochlea, esophagus, eyes, mandible, oral cavity,<br>various lymph nodes Breast region: Body, heart, trachea, various<br>lymph nodes Pelvis: Body, bladder, femoral heads, pelvic<br>bones, rectum, spinal canal The system calculates the anatomical points, image<br>features, and similarity scores of the patient image and<br>compares them with pre-stored expert cases. Rigid<br>registration is used both for initializing the deformable<br>registration algorithm and for displaying the expert case<br>and patient image in aligned preview. The<br>autocontouring approach depends on the selected<br>structures. Either the structures are heuristically<br>segmented from the patient image, or the structures are<br>generated via deformable registration and structure<br>propagation from the expert cases. If multiple expert<br>cases are used, the propagated structures from the<br>different atlases are fused by means of the simultaneous<br>truth and performance level estimation (STAPLE)<br>algorithm. | The RPA uses deep learning algorithms<br>which Eclipse does not.<br>Use and function: In Eclipse, the user edits<br>the contours prior to planning. This is the<br>same for complex planning in the RPA<br>(VMAT planning for head / neck and cervix).<br>It is different for simple plans, where the plan<br>is generated before the user reviews the<br>contours. If the user edits the contours in the<br>RPA, they will have to delete the plan as<br>well. | Use and function: The RPA<br>provides autocontouring for a<br>range of structures, including<br>most of those listed here for<br>Eclipse.<br>Performance data: The<br>algorithm for Expert<br>Segmentation in Eclipse is<br>very similar to the Multi-Atlas<br>Contouring System (MACS)<br>that is used to contour<br>structures for the chest wall<br>planning in the RPA.<br>Safety and effectiveness: Both<br>Eclipse and the RPA are<br>designed to provide contours<br>that the users review and edit. | Devices are Substantially<br>Equivalent. Both devices<br>provide autocontouring<br>functions for the same<br>anatomical regions. Both<br>devices require user edits of<br>contours with 'complex plans'<br>prior to planning. |
| Eclipse,<br>Other Plan<br>Preparation | Automatic marker detection - Eclipse v.15.6 includes a<br>function ('Calypso Beacon Detection') to automatically<br>detect a specific type of marker (Calypso transponders)<br>on CT images. | Use and function: The Eclipse function is for<br>a specific type of marker that is different<br>from the generic markers that the RPA is<br>designed for. | Use and function: Both Eclipse<br>and the RPA can automatically<br>detect markers. | Devices are Substantially<br>Equivalent. Both devices can<br>automatically detect markers. |
| Comparison of the RPA System's Software Functions with the | | | | |
| Eclipse Treatment Planning System Version 15.6 (K181145) | | | | |
| Software | Description of Functions Available in | Differences | Similarities | Rationale for Substantial |
| Function | Eclipse | | | Equivalence |
| Eclipse<br>Automated<br>Planning,<br>VMAT | 1. Photon Optimizer (PO) algorithm. This algorithm<br>is used to optimize IMRT or VMAT plans based on<br>DVH constraints / objectives.<br>2. Automated Optimization Workflow. Enabling this<br>can automate the optimization workflow for IMRT<br>planning so that, after optimization, the leaf motion<br>calculation and final dose calculation are<br>automatically initiated, and the results are then<br>automatically saved. A similar feature exists for<br>VMAT plans.<br>3. DVH Estimation Models for RapidPlan. DVH<br>estimation models are created from information<br>extracted from a set of previous treatment plans<br>(called 'treatment plans'). The estimation models<br>predict the DVH that is achievable from the current<br>treatment plan (based on the geometry in the current<br>plan), and also creates a set of optimization objects<br>that can be based on the DVH estimates or fixed<br>(i.e., not based on the DVH estimates). | Use and function: The main difference for<br>VMAT planning is that Eclipse generally<br>creates a plan that the user reviews, makes<br>edits to the optimization constraints, and<br>repeats the process to improve the plan<br>quality. The RPA uses the same<br>optimization tools (i.e., the tools in Eclipse),<br>but the optimization objectives and<br>constraints have been pre-set to give optimal<br>plans for the majority of patients. The user<br>is not able to easily edit the RPA VMAT<br>plans so, if they do not approve the plan for<br>clinical use, they must delete it and create<br>one using their own routine processes (i.e.,<br>in their own treatment planning system). | Use and function: The RPA<br>uses some Eclipse features,<br>including DVH Estimates for<br>RapidPlan and the Photon<br>Optimizer for optimizing<br>VMAT plans. The plans look<br>very similar.<br>Safety and effectiveness: Both<br>Eclipse and the RPA are<br>designed to create plans that<br>the users then edit and review<br>for clinical acceptability prior<br>to use. | Devices are Substantially<br>Equivalent. Both devices provide<br>autoplanning features and create<br>plans that the users then edit and<br>review for clinical acceptability<br>prior to use. Both devices provide<br>Photon Optimizer, automated<br>optimization workflow and DVH<br>estimation models. |
| Autoplanning,<br>Other | 1. Beam Angle Optimization (BAO). This tool<br>optimizes the number and angle of treatment beams.<br>It optimizes the objective function, which is<br>determined by DVH goals / constraints and a normal<br>tissue objective (which falls off with distance from<br>the PTV). BAO can be used for IMRT plans or as a<br>starting point for conformal treatment plans.<br>2. Collimator Angle Optimization (CAO). This<br>function optimizes collimator angle for each arc of a<br>HyperArc plan such that, whenever possible, a given<br>pair of MLC leaves delineates only one target in the<br>beam's-eye-view.<br>3. Optimize collimator jaws. Adjusts the collimator<br>jaws to best fit the MLC leaves to the structure.<br>4. Use recommended jaw positions. Adjusts the<br>collimator jaw positions with an additional margin.<br>5. Optimize collimator rotation. Optimizes the<br>collimator rotation around a structure. | Use and function: Autoplanning in Eclipse<br>is mostly automation of individual tasks that<br>are controlled by the user. The user does<br>not control these tasks with the RPA.<br>Use and function: Review and editing of 3D<br>plans (cervix 4-field box, post-mastectomy<br>breast plans, whole brain plans) for the RPA<br>happens in the users' own treatment<br>planning system. | Use and function: Many of the<br>treatment plan details in<br>Eclipse and RPA use functions<br>with similar algorithms, such<br>as optimizing the jaw<br>positions.<br>Safety and effectiveness: Both<br>Eclipse and the RPA are<br>designed to create plans that<br>the users then edit and review<br>for clinical acceptability prior<br>to use. | Devices are Substantially<br>Equivalent. Both devices create<br>plans that the users then edit and<br>review for clinical acceptability<br>prior to use through the use of AI<br>software. Both devices provide<br>Beam Angle Optimization,<br>Collimator Angle Optimization and<br>collimator jaw optimization. |

### Table 2: Comparison of Functions of the Subject Device with Functions of the Predicate Device

{7}------------------------------------------------

{8}------------------------------------------------

### PERFORMANCE DATA 7.

#### 7.1 Non-Clinical Data

#### 7.1.1 Software Verification and Validation Testing

Software verification and validation was conducted, and documentation was provided as recommended by the FDA's Guidance for Industry and FDA Staff, "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices." The software for this device was considered as a "major" level of concern. Test results demonstrate conformance to applicable requirements and specifications.

No animal studies or clinical tests have been included in this pre-market submission.

The ground truth treatment plans were generated by the primary 4-field box automation technique for cervical cancer by Kisling et al. (Kisling 2019) with beam apertures based on a patient's bony anatomy. Only the clinically acceptable plans were used for training (rated by physicians); their DRRs and corresponding beam apertures were the inputs for training (and just the DRRs for testing/prediction). No additional criteria were applied. The test set was generated in the same manner as the ground truth, but on previously unseen patients.

Initial software training for each anatomical location was successfully accomplished and is described in brief in Table 3 below.

{9}------------------------------------------------

Table 3 presents the initial testing performed for software training and testing. Multicenter performance testing is presented in Section 7.2.

| Anatomical<br>Location | Tissue Type(s) | Training Data Set | Test Data<br>Independence |
|------------------------|------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Head and Neck | Normal Tissue<br>(primary) | 3,288 patients (3,495 CT scans) who received radiation therapy at MD<br>Anderson Cancer Center between September 2004 and June 2018. Any<br>patient who received a simulation CT scan of the head/neck region in a<br>head -first supine position was eligible. | 174 CT scans were<br>randomly selected from<br>this group (and excluded<br>for training) plus<br>qualitative evaluation 24<br>CT scans from an<br>external dataset. |
| Head and Neck | Normal Tissue<br>(secondary) | 160 patients who received radiation therapy at MD Anderson Cancer<br>Center from 2018 to 2020. Any patient who received a simulation CT<br>scan of the head/neck region in a head-first supine position was eligible. | Test patients were<br>randomly selected and<br>excluded from the<br>training set. |
| | Lymph Node<br>CTVs | 61 patients who received radiation therapy at MD Anderson Cancer<br>Center between 2010 and 2019. Any patient who received a simulation<br>CT scan of the head/neck region in a head-first supine position was<br>eligible. | These 71 cases were<br>randomly placed in 3<br>groups: training (51 pts.),<br>cross-validation (10 pts.)<br>and final test (10 pts.). |
| Whole Brain | Whole Brain | The whole brain primary segmentation models used the same models as<br>used for head and neck segmentation, described above, as well as an<br>additional vertebral body localization and segmentation model (Vertebral<br>Bodies model: spinal canal CNN: 1,966, VB labeling: 803, VB<br>segmentation: 107, from 930 MDACC patients and 355 external<br>patients). Patients who received spinal radiotherapy for spinal metastases<br>(3DCRT and VMAT) at MD Anderson, or for whom data was publicly<br>available (MICCAI challenge data). | Test patients were<br>randomly selected from<br>this group (and excluded<br>for training). |

Table 3: Software Training for Anatomical Locations

{10}------------------------------------------------

| Anatomical<br>Location | Tissue Type(s) | Training Data Set | Test Data<br>Independence |
|------------------------|---------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| GYN | Normal Tissue<br>(primary) | 1,999 patients (2,254 CT scans) who received radiation therapy at MD<br>Anderson from September 2004 and June 2018. Any patient who<br>received a simulation CT scan of the pelvic region in a head-first supine<br>position was eligible. | 140 CT scans were<br>randomly selected from<br>this group (and excluded<br>for training) plus<br>qualitative evaluation<br>with 30 cervical cancer<br>patients from 3 centers in<br>S. Africa. |
| GYN | Normal Tissue<br>(secondary) | 192 patients (316 CT scans) who were treated for locally advanced<br>cervical cancer between 2006 and 2020. | Test patients were<br>randomly selected from<br>this group (and excluded<br>for training). |
| GYN | CTVs (primary) | 406 CT scans from 308 patients (UteroCervix), 250 CT scans from 201<br>patients (Nodal CTV), 146 CT scans from 131 patients (PAN), 490 CT<br>scans from 388 patients (Vagina), 487 CT scans from 388 patients<br>(Parametria) who received radiation therapy at MD Anderson Cancer<br>Center between 2006 and 2020. | Test patients were<br>randomly selected from<br>this group (and excluded<br>for training). |
| | Liver | Training data for GYN Liver (normal) comprised 119 patients (169 CT<br>scans) who had received contrast-enhanced and non-contrast CT imaging<br>of the liver at MD Anderson Cancer Center. | Test patients were<br>randomly selected from<br>this group (and excluded<br>for training). |
| Chest Wall | Whole Body<br>(secondary for<br>chest wall) | Training data for whole body (secondary for chest wall) comprised 250<br>patients who were treated at MD Anderson between August 2016 and<br>June 2021, with CT imaging in the thoracic region. | Test patients were<br>randomly selected from<br>this group (and excluded<br>for training). |

{11}------------------------------------------------

#### Standards Conformance 7.1.2

The subject device conforms in whole or in part with the following standards:

- IEC 62304 Medical device software Software life cycle processes
- IEC 62083 Requirements for the safety of radiotherapy treatment planning systems

#### 7.2 Clinical Data

A summary of the multicenter clinical data is presented in the tables below.

{12}------------------------------------------------

| Characteristic | All<br>Cervix<br>VMAT | All<br>Cervix<br>3D | All Chest<br>Wall | All Head<br>and Neck | All<br>Whole<br>Brain |
|----------------------------------------------|-----------------------|---------------------|-------------------|----------------------|-----------------------|
| No. of Unique Patients with RPA Plan(s) | 50 | 47a; 45b | 46 | 86 | 46 |
| CT Scan Equipment | | | | | |
| Philips | x | x | x | x | x |
| Siemens | x | x | x | x | x |
| GE | x | x | x | x | x |
| No. of Clinical Sites | 5 | 5 | 5 | 5 | 5 |
| No. of Participating Physicians / Study Site | | | | | |
| Site 1 | 5 | 5 | 8 | 12 | 12 |
| Site 2 | 1 | 1 | 2 | 3 | 2 |
| Site 3 | 3 | 3 | 1 | 1 | 3 |
| Site 4 | 3 | 3 | 8 | 1 | 6 |
| Site 5 | 2 | 2 | 4 | 5 | 2 |
| Clinical Subgroups and Confounding Factors | | | | | |
| By Study Site | None | None | None | None | None |
| By Equipment | None | None | None | None | None |
| Age | | | | | |
| Mean | 51 | 50 | 51 | 62 | 60 |
| Min, Max | 26, 94 | 26, 84 | 31, 80 | 27, 87 | 14, 88 |
| Sex | | | | | |
| Male | 0.0% | 0.0% | 2.2% | 79.3% | 39.1% |
| Female | 100.0% | 100.0% | 97.8% | 29.7% | 34.8% |
| Not Reported | 0.0% | 0.0% | 0.0% | 0.0% | 26.1% |
| Race | | | | | |
| Asian | 9.8% | 2.1% | 26.1% | 5.4% | 6.5% |
| Black/African American | 13.7% | 14.9% | 13.0% | 12.0% | 8.7% |
| White | 39.2% | 78.7% | 32.6% | 73.9% | 54.3% |
| Native Hawaiian or Pacific Islander | 0.0% | 0.0% | 0.0% | 1.1% | 0.0% |
| British | 7.8% | 0.0% | 0.0% | 0.0% | 0.0% |
| American Indian or Alaskan Native | 2.0% | 0.0% | 4.3% | 1.1% | 0.0% |
| Other / not available | 27.5% | 4.3% | 23.9% | 6.5% | 28.3% |
| Ethnicity | | | | | |
| Hispanic or Latino | 25.5% | 10.6% | 8.7% | 7.6% | 6.5% |
| Not Hispanic or Latino | 41.2% | 46.8% | 43.5% | 50.0% | 58.7% |
| Other / not available | 33.3% | 42.6% | 47.8% | 42.4% | 34.8% |

| Table 4: | | Demographics, Number of Patients, Number of Samples, and Clinical Sites |
|----------|--|-------------------------------------------------------------------------|
| | | |

ª4-field box soft tissue plan

b4-field box bony landmark plan

{13}------------------------------------------------

| Criteria<br>Number | Criteria | Results |
|--------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------|
| 1 | Assess the safety of using the RPA plan for normal structures for treatment<br>planning by comparing the number of patient plans that pass accepted<br>dosimetric metrics when assessed on the RPA contour with the number that<br>pass when assessed on the clinical contour. The difference should be 5% or<br>less. When there are multiple metrics for a single structure at least one<br>should pass this criterion. | < 5% difference<br>between RPA Plan and<br>Clinical Plan for all<br>bony structures and<br>critical soft tissue<br>structures with VMAT<br>and 4 field box.* |
| 2 | Assess the effectiveness of the RPA plan for normal structures by<br>comparing the dose to RPA normal structures for RPA plans and clinical<br>normal structures for clinical plans. The difference in the number of RPA<br>plans that pass accepted dosimetric metrics and the number of clinical plans<br>that pass accepted dosimetric metrics should be 5% or less. When there are<br>multiple metrics for a single structure at least one should pass this criterion. | < 5% difference<br>between RPA Plan and<br>Clinical Plan for all<br>bony structures and<br>critical soft tissue<br>structures with VMAT<br>and 4 field box.** |
| 3 | Assess the effectiveness of the RPA plan for target structures by comparing<br>the number of RPA plans that pass accepted dosimetric metrics (e.g.,<br>percentage volume of the PTV receiving 95% of the prescribed dose) when<br>compared with clinical plans. The difference should be 5% or less. When<br>there are multiple metrics used to assess a single structure, at least one<br>coverage and one maximum criterion should pass this criterion. | < 5% difference<br>between RPA Plan and<br>Clinical Plan for all<br>assessed structures |
| 4 | Assess the geometric effectiveness of the RPA targets using recall. A low<br>value for this metric represents under-contouring. The 25th percentile of the<br>recall must be 0.7 or greater. | 25th percentile for<br>recall > 0.7 |
| 5 | Assess the quality of body contouring generated by the RPA by comparing<br>primary and secondary body contours generated by the RPA with manual<br>body contours. Surface DSC (2mm) should be greater than 0.8 for 95% of<br>the CT scans. | Surface DSC > 0.8 for<br>95% of CT scans |
| 6 | Assess the ability of the RPA to accurately identify the marked isocenter.<br>This is achieved by comparing the automatically generated isocenters with<br>manually generated ones. 95% of automatically generated marked<br>isocenters (primary and verification approaches) should agree with<br>manually generated marked isocenters within 3mm in all orthogonal<br>directions (AP, lateral, cranial-caudal). | < 3mm difference<br>between RPA Plan and<br>Clinical Plan for all<br>orthogonal directions |

Table 5: Summary of Statistical Results—Cervix

* With the exception of bowel bag in the 4-field box plans, the RPA contour gives a more conservative result. ** RPA plan and clinical plan had 6% - 13% difference in passing rates using VMAT on right kidney, bladder, and bowel. The RPA Plan for rectum exceeded passing rates of the clinical plans in excess of 5%. However, when the RPA plan (which was created using the RPA normal contours) was assessed using the clinical normal contours, the passing rates for the clinical plan and RPA plan are within 5% for all normal structures. This is a result of the conservative nature of the RPA contours.

{14}------------------------------------------------

| Criteria<br>Number | Inclusion Criteria | Exclusion Criteria | Sampling Method |
|--------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------|
| 1 | CT scan of the female pelvic anatomy. | Poor Image Quality | |
| 2 | Clear CT image of the pelvic region without<br>distortions. | - | |
| 3 | Test datasets consisted of CT images of<br>patients previously treated for cervical<br>cancer using radiotherapy following one of<br>the following treatment schemes:<br>• 4-field box (based on bony<br>landmarks or soft tissue)<br>• VMAT | - | |
| 4 | Scan was obtained with patient head-first,<br>supine. | - | Test datasets were chosen<br>going forward in time<br>until sufficient data were |
| 5 | The datasets included CT images, original<br>clinical contours of anatomic structures and<br>treatment targets, and the dose distributions<br>used for patient treatment. | - | collected, starting with CT<br>scans collected on January<br>1, 2022. If insufficient<br>patient scans were found, |
| 6 | Test datasets were chosen going forward in<br>time until sufficient data was collected,<br>starting with CT scans collected on January<br>1, 2022. If insufficient patient scans were<br>found, data collection could be restarted with<br>January 1, 2021 (for patients treated in 2021)<br>and so forth, until sufficient data was<br>collected. | - | data collection was<br>restarted with January 1,<br>2021 (for patients treated<br>in 2021) and so forth, until<br>sufficient data was<br>collected. |
| 7 | Testing datasets were unique, with no<br>overlap with data used for model creation or<br>in previous validation studies. | - | |
| 8 | CT scans included the manufacturer and<br>model of the scanner used to obtain the CT<br>image. | - | |

Table 6: Summary of Cervix Protocol

{15}------------------------------------------------

| Criteria<br>Number | Criteria | Results |
|--------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------|
| 1…