ASTRO 20: RefleXion Presents New Research on Biology-Guided Radiotherapy for Metastatic Cancer

By News Release

RefleXion Medical, a therapeutic oncology company pioneering biology-guided radiotherapy* (BgRT) as a new modality for treating all stages of cancer, has showcased eight clinical abstracts evaluating the use of its novel technology at the virtual American Society for Radiation Oncology (ASTRO) 2020 Annual Meeting, Oct. 24-28.

“We are thrilled to have seven abstracts from leading academic clinicians highlighting our technology,” said Sean Shirvani, M.D., senior vice president of Clinical and Medical Affairs at RefleXion. “The presentations will share new data on the potential of BgRT for targeting tumors in diverse anatomic locations and from varied histologies, including non-small cell lung, prostate and pancreatic cancers. The investigations also explore the exciting opportunity of leveraging PET information as a future biomarker with BgRT.

“Through our clinical partnerships, we continue to produce comprehensive, evidence-based research that builds on the strong foundation of previously published studies,” continued Shirvani. “We believe the RefleXion X1 can form the backbone of any radiotherapy clinic, as it delivers IMRT/SBRT/SRS and unlocks opportunities for procedural growth with BgRT in local and metastatic disease.” 

The abstracts are as follows:

Feasibility of Biology-guided Radiotherapy (BgRT) Targeting Fluorodeoxyglucose (FDG) Avid
Liver Metastases

A Amini, D Du, T Abuali, J Neylon, D Zuro, SM Shirvani, C Huntzinger, A Da Silva, S Hui, J Wong,
A Liu City of Hope Cancer Center, Duarte, CA


Biology-guided radiotherapy (BgRT) is a novel RT delivery approach utilizing fluorodeoxyglucose (FDG)
activity on positron emission tomography/computerized tomography (PET/CT) scans performed in real-
time to track and direct RT. An adequate contrast in FDG activity between the tumor and the background
tissue, referred to as the normalized SUV (NSUV), is necessary for optimal functioning of BgRT. Organs
such as the liver have some background FDG activity posing some potential challenges for BgRT. This
study seeks to characterize the NSUV in metastatic lung adenocarcinoma patients with liver metastases.


We reviewed the charts of 50 lung adenocarcinoma patients with liver metastases diagnosed from 20162018 at our institution. Of those, 27 patients had PET/CT scans with FDG avid liver metastases. VelocityTM oncology imaging informatics system was utilized to capture SUV values. A threshold SUV of 4 was selected a priori to auto contour each liver metastasis. The following variables were collected: SUVmax and SUV mean for each liver metastasis, SUVmean and SUVmax 5 mm radially from lesion, and SUVmean and SUVmax 10 mm radially from lesion. SUV background of liver was captured, NSUV at 5 mm (SUVmax of liver metastasis divided by SUV mean at 5 mm), and NSUV at 10 mm (SUVmax of liver
metastasis divided by SUV mean at 10 mm) was calculated for each metastasis.


In total, 27 patients were included in the final analysis, with 82 measurable lesions. The average
SUV background of liver was 2.26 (95% confidence interval [CI] 2.17-2.35); the average SUV mean for
liver metastases was 5.31 (95% CI 4.87-5.75) and the average SUVmax of liver metastases was 9.19 (95% CI 7.59-10.78). The average SUV mean at 5 mm and 10 mm radially from each lesion was 3.08 (95% CI 3.00-2.16) and 2.60 (95% CI 2.52-2.68) respectively. The mean NSUV at 5 mm and 10 mm were 3.46 (95% CI 2.21-4.71) and 4.08 (95% CI 2.61-5.54) respectively. Furthermore, 90% of lesions had NSUV greater than 1.93 at 5 mm and 2.28 at 10 mm.


This is the first study to comprehensively characterize tumor-to-background SUV contrast in liver
metastases in the context of BgRT. This evidence will guide optimization of BgRT use for treating liver
metastases given the presence of background SUV normally found in the liver.

Feasibility of Biology-Guided Radiotherapy for Pancreatic Tumors: An Assessment of NormalizedTarget SUV

RR Patel, Tinsu Pan, SM Shirvani, C Huntzinger, A Da Silva, V Verma, A Koong, E Koay, JW Welsh, The University of Texas MD Anderson Cancer Center, Houston, TX, USA

Biology-guided radiotherapy (BgRT) is a novel method of delivering radiation using emissions from
injected radiotracers to provide tracked radiotherapy. Pancreatic tumors in particular may benefit from
BgRT given the potential for reducing irradiated volumes and thereby reducing gastrointestinal toxicity.
For BgRT to be feasible, there must be sufficient contrast between tumor targets and the surrounding
background. We assessed pancreatic lesion contrast using normalized target standardized uptake values
(NSUVs) calculated from FDG-PET scans.

Patients were identified with IRB approval from our institution (2019-1073). We measured NSUVs on
scans from patients with pancreatic cancer before and after induction chemotherapy, which in most cases
was FOLFIRINOX or gemcitabine with nab-paclitaxel. Tumors were contoured on PET/CT scans. We
calculated the average SUV for two margins around the gross tumor volume (GTV) that did not include the
GTV: one for a 5-mm margin and the other for a 10-mm margin. NSUVs were calculated as the SUVmax
of the GTV divided by the average SUV of the margin, and then averaged across axial, coronal, and sagittal views. SUVs were measured retrospectively on a GE AW Server 3.2. Since not all PET/CT scans were done with intravenous contrast, the major limitation of this study was identifying the GTV in those patients. Non-parametric Mann-Whitney U and Kruskal-Wallis tests were used to compare NSUV before and after chemotherapy and by tumor location.

We scanned tumors in 53 patients with pancreatic adenocarcinoma, 29 at baseline and 28 after induction
chemotherapy; 4 patients had scans available at both times. No patient had received pancreatic surgery or
radiotherapy. Most tumors were located in the head of the pancreas (33) followed by the body (9), tail (6),
neck (3), and uncinate process (2). Before therapy, the SUVmax range was 8.64‒72.25; the mean NSUV
was 3.84 (range 1.77‒7.33) for 5-mm margins and 4.05 (1.77‒7.91) for 10-mm margins. After therapy, the
SUVmax range was 5.13‒57.43; the mean NSUV was 3.02 (range 1.21‒9.70) for 5-mm margins and 3.28
(1.21‒8.67) for 10-mm margins. Decreases in NSUV after chemotherapy were significant for both the 5mm
(p=0.001) and 10-mm (p=0.01) margins. For the 4 patients with evaluable scans before and after
induction chemotherapy, the NSUV after induction chemotherapy changed by ‒1.15 for 5-mm margins and
by ‒1.12 for 10-mm margins. No differences were observed in terms of tumor location (p>0.05). Although exact NSUV threshold criteria are yet to be defined, using thresholds of 2.5, 3, or 3.5 and a 10mm margin, 86%, 72%, and 59% of pre-therapy patients and 68%, 50%, and 29% of post-induction chemotherapy
patients, respectively, would be considered suitable for BgRT.

BgRT may be feasible in patients with pancreatic cancer.

Suitability of PSMA-PET Biology-guided Radiotherapy for Low Volume Metastases in Newly
Diagnosed Prostate Cancer

M Gaudreault, N Hardcastle, P Jackson, J Callahan, T Kron, C. Huntzinger, SM Shirvani, A Da Silva, MHofman, GG Hanna, & Shankar Siva, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia


Biology-guided radiation therapy (BgRT) directs radiation therapy in real time based on positron emissions
detected during treatment delivery. Protein specific membrane antigen (PSMA) is a positron emission
tomography (PET) tracer with superior sensitivity and specificity for prostate cancer detection compared
with conventional imaging. This study aims to quantify the ratio of standard uptake value (SUV) to
background SUV for patients with low volume prostate cancer metastases, in the context of determining
suitability of these lesions for BgRT.


A single institutional patient subset from the ProPSMA prospective clinical trial (NCT trial ID
ACTRN12617000005358) underwent Ga-68-PSMA-11 at the time of prostate cancer diagnosis with
suspected localised disease. From this cohort of 84 patients, 15 had at least one metastatic site disease
diagnosed, 1 to 11 lesions per patient, with a total of 43 metastatic lesions (12 bones/31 viscerals).
Metastatic gross tumour volume (GTV) were delineated on PSMA-PET images by using a semi-automated threshold method. In order to determine the suitability of various BgRT tracking zones, 3D shells of thickness 5 mm, 10 mm and 20 mm were constructed around the GTV to represent adjacent non-tumour background. Normalized SUV (nSUV) was calculated as the ratio of SUVmax in the GTV to SUVmean ofadjacent shell. Sensitivity of nSUV to shell thickness variation was reported as the difference between nSUV values obtained for each thickness.


GTV SUVmax ranged from 2.9 to 48.0 (median, 12.0; IQR, 5.1 to 19.5). nSUV ranged from 2.4 to 42.6 for
5 mm shell, 2.7 to 47.5 for 10 mm shell and 2.6 to 36.4 for 20 mm shell. Bone metastases had increased
SUVmax compared with visceral metastases (SUVmax min/median/max 2.9/15.0/48.0 for bone compared
with 2.9/11.0/40.0 for visceral). Furthermore, 90% of lesions had nSUV greater than 2.9 for a shell
thickness of 5 mm, 3.1 for a shell thickness of 10 mm, and 3.2 for a shell thickness of 20 mm. Moreover,
100%/93%/84% of lesions would be suitable for BgRT by using nSUV threshold of 2/3/4 and a shell
thickness of 10 mm. nSUV decreased with increasing shell thickness for 17 lesions, indicating high signal
in adjacent non-tumour tissue due to the proximity of bladder (5), bowel (4), other lesion (4), vessel (3) andliver (1).


This study demonstrates that it is feasible to identify metabolically defined radiotherapy targets in the
setting of BgRT. High lesion to background signal was observed for metastases visible on Ga-PSMA-EPT11; however, a subset of metastases had adjacent non-tumour uptake, which may require exclusion fromemission tracking during BgRT.

Simultaneous Integrated Boost of Lung Tumors in the Stereotactic Ablative Setting using BgRTTracked Delivery

P Olcott, SM Shirvani, S Tian, I Sethi, X Yang, A Da Silva, C Huntzinger, S Mazin1, TK Owonikoko,
DM Schuster, WJ Curran, KA Higgins Emory University Cancer Center, Atlanta, GA
RefleXion Medical, Hayward, CA


Biology-guided radiotherapy is a new radiation modality that utilizes real-time partial FDG images of
tumors to deliver a dynamically tracked dose distribution. A key application is to ablate moving lung tumors
without prescribing the therapeutic dose to the entire motion envelope of the tumor as is done with internal tumor volume (ITV) techniques. Better normal tissue sparing with a BgRT tracked dose distribution may furthermore enable delivery of a simultaneous integrated boost (SIB) to the tracked GTV to improve tumor response. This study assesses the utility of the BgRT SIB approach for mobile lung tumors and compares resultant dosimetric parameters with SBRT-ITV technique.


Two representative cases of lung tumors with meaningful motion profiles that previously underwent lung
SBRT were selected for this BgRT planning study. The planning objective was to deliver 50 Gy to the
planning target volume (PTV) and 60 Gy simultaneous integrated boost to the gross tumor volume (GTV)
in 5 fractions. In BgRT, a single respiratory phase was used to contour the tracked GTV and a 5 mm
expansion was used to delineate the tracked PTV. For SBRT, prescription volumes were contoured by
summing target volumes across all respiratory phases to create an internal planning tumor volume (IPTV)
and an internal gross tumor volume (IGTV). All SBRT and BgRT plans were generated on a prototype
BgRT treatment planning system.


Both tumors had maximum excursions of 1.2 cm. The BgRT SIB plans met all dosimetric objectives and
normal tissue constraints. Maximum achieved GTV doses were 80 Gy and 74.4 Gy in the two patients,
while corresponding lung V20s were 4.2% and 2.8%, respectively. The BgRT-PTV was 55.9% and 40.7%
smaller in the two patients when compared to the corresponding SBRT-IPTV. Likewise, the BgRT-GTV
prescription volume was reduced by 68% and 43.3%, respectively, when compared to the corresponding
SBRT-IGTVs. Compared to SBRT-ITV plans, BgRT plans resulted in average relative percentage reductions in all lung dosimetric parameters; the lung V20, V10, V5, D1500cc, and MLD were reduced by 20.7%, 15.1%, 15.3%, 30% and 19.0%, respectively.


BgRT plans resulted in reduced target volumes and better organ sparing compared to SBRT in the context of SIB to a moving lung target. This investigation provides proof-of-concept that dose distributions
delivered by dynamic tracking based on partial FDG images can enable dose escalation to the GTV in the
stereotactic ablative setting. This approach may be particularly beneficial for radioresistant lung metastases, such as those arising from colorectal or sarcoma primary tumors.

Increased 18-FDG Metabolic Activity During Lung SBRT Predicts Risk of Disease Progression:
Results from a Prospective Study of Serial Inter-Fraction PET/CTs

S Tian, JM Switchenko, X Yang, I Sethi, A Da Silva, TK Owonikoko, DM Schuster, WJ Curran, KA, Emory University Cancer Center, Atlanta, GA

Static measures of metabolic activity, based on pre-or post-treatment PET/CTs have informed response to treatment and prognosis in lung cancer. Whether inter-fraction changes in metabolic parameters during lung SBRT can predict disease outcomes is unknown.

Patients treated with lung SBRT of 50 Gy in 5 fractions for early-stage NSCLC or lung metastases
underwent 3 PET/CTs per protocol. Studies were acquired within 2 weeks of treatment start (PET1),
between fractions 1 and 2 (PET2), and fractions 4 and 5 (PET3). FDG-PET parameters including maximum and mean standardized uptake value (SUVmax, SUVmean), metabolic tumor volume (MTV), with additional normalized and derived metrics were extracted. The effect of static and change/dynamic metrics on the risk of local and distant progression were determined for each PET parameter. Continuous and categorical variables were compared by non-parametric methods via Fisher’s exact and Kruskal-Wallis
tests. Treatment-related adverse-effects (AE) occurring within 30 days (acute) and 12 months (late) of
SBRT were graded via CTCAE v5.0.

14 patients, who completed all protocol-directed imaging and treatment, representing 17 target lesions,
were included for analysis. Median follow-up length was 16.7 months. 7 patients received SBRT as
definitive treatment for early-stage disease, 7 received SBRT in the recurrent/metastatic setting. Local tumor failure occurred at 2 treated lesions; actuarial local (tumor) control was 93.3% at 1-year and 80.8% at 2 years. 4 patients developed distant disease progression; freedom from distant progression was 85.7% at 1 year, 24.9% at 2 years. Lung SBRT with intra-fraction PET/CT was well-tolerated. A single patient developed a grade 2 AE (odynophagia) in the acute setting. A single patient developed a G3 AE
(pneumonitis), and 2 developed G2 events (fatigue, chest wall pain). There were no toxicities attributed to
serial PET/CTs. Use of serial PET/CTs identified multiple PET predictors of disease progression. In
recurrent/metastatic patients, a greater increase in lesion SUVmax from PET2 to PET3 (PET2-3) was
associated with distant failure (mean change -0.65 vs 7.2, p=0.025). Increases in SUVmax/liver SUV mean
were similarly linked to distant progression for both PET1-3 and PET2-3 (p=0.025). Increases in 4 other
dynamic PET variables were identified as significant predictors for distant failure in patients with
recurrent/metastatic disease: PET1-3 SUVmax/SUVmean, PET2-3 SUVmean/liver SUVmean, PET1-2
MTV/GTV, and SUVmax/mean coefficient of variation. 2 additional variables were identified in the early-
stage cohort, and 4 variables when both groups were analyzed in aggregate. Notably, no static PET variables were significant predictors of subsequent local or distant failures.

Rising metabolic activity during lung SBRT are potential markers of distant disease progression, as
demonstrated by multiple PET parameters acquired via serial 18-FDG PET/CTs.

Use of a Detailed Process Map for Clinical Workflow of a New Biology-guided Radiation Therapy

MS Hwang , RJ Lalonde, & S Huq. University of Pittsburgh School of Medicine, Pittsburgh, PA


A newly proposed clinical technology for biology-guided radiation therapy (BgRT) uses PET emissions to track and treat multiple metastatic tumors in real-time. The use of PET and radiopharmaceutical tracers for treatment introduces new processes that will need new guidelines to implement the technology safely.
In this study, we aimed to understand and develop a clinical workflow for staff and patients using a prospective process map to implement the prototype BgRT machine. It will be used to instruct the new
therapeutic strategy when the BgRT machine becomes available for clinical use.


A team of 16 members from various radiation therapy disciplines at our institution actively participated in
developing a prospective process map for a prototype BgRT machine. The steps unique to their workflow were analyzed by the team members in charge and then assigned accordingly to a responsible member. The process map was developed for the case of a radiation oncology department already equipped with a PET-CT system. The target goal for BgRT is to treat multiple metastases in the same treatment session found in areas including head & neck, lung, liver, and lymph nodes using stereotactic body radiation therapy (SBRT) with fluorodeoxyglucose radiolabeled with fluorine-18 (FDG).


The process map consists of a total of 15 sub-processes and 131 steps. Seven sub-processes are unique to BgRT, originating from acquiring multiple PET images at the diagnostic PET-CT unit and the BgRT
machine with separate patient visits, creating a unique biological treatment volume for BgRT plan
(PTVBgRT) and a typical PTVSBRT for standard SBRT plan, use of an external server for registration/contouring, and the BgRT plan optimization and treatment delivery using PET images. Of the131 steps, the radiation oncologists (RO) are primarily responsible for 26 steps (19.8 %), the medical physicist (MP) for 39 steps (29. 8 %), the nuclear medicine technologist (NMT) for 24 steps (18.3 %), the radiation therapist (RT) for 39 steps (29.8 %), and 3 steps (2.3 %) for initial patient record and evaluation.


As expected, the treatment of multiple metastases in the same session will impact clinical workflow, thus our desire to create a graphical process map depicting the new clinical workflow with an appropriate level of detail. We have identified the steps unique to this new clinical process, the relationships between steps,
and have highlighted the possible risks or complexity caused by the new steps. The minimum staffing
requirement in each profession, (e.g., RO, MP, NMT and RT), and maximally allowed daily BgRT patients are yet to be determined. Staff training/education to meet regulatory requirements, in particular for the RO as an authorized tracer user, needs to be considered. Strong cooperation with a local PETtracer vendor and constructing patient schedules to allow for multiple time-sensitive PET images at the treatment machine are also keys to the new BgRT workflow.

Prognostic Value of FDG-PET Metrics for Advanced NSCLC Patients Treated with First-line Immunotherapy

TY Andraos, B Halmos, H Cheng, C Huntzinger, & N. Ohri, Albert Einstein College of Medicine and Montefiore Medical Center, Bronx, NY


Measures of disease burden on 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) have
prognostic value for advanced non–small cell lung cancer (NSCLC) patients treated with chemotherapy.
Here we explore the prognostic value of FDG-PET for patients treated with first-line immunotherapy and
if PET-guided radiotherapy (RT) should be explored in those patients.


Patients treated at our institution between 2015 and 2019 with first-line immune checkpoint inhibitors
targeting the PD-1/PD-L1 axis (with or without chemotherapy) for advanced/metastatic NSCLC who
underwent PET staging were included in this analysis. A commercially available gradient-based
segmentation tool was used to contour all visible hypermetabolic extracranial lesions on each staging PET.
Number of hypermetabolic lesions (nLesions) and total metabolic tumor volume (MTV, log-transformed)
were tested as predictors of progression-free survival (PFS) and overall survival (OS) duration in univariate Cox proportional hazards models. Among patients who received RT near the time of diagnosis, the proportion of disease treated with RT was tested as prognostic factor as well. Actuarial PFS and OS rates were calculated using the Kaplan-Meier Method, and comparisons between subgroups were performed using logrank testing.


Seventy-two patients met inclusion criteria. Thirty-two patients (44%) received immunotherapy as
monotherapy, and 40 (56%) received immunotherapy with chemotherapy. PD-L1 tumor proportion score(TPS) was below 50% in 38 patients, 50-100% in 24 patients, and unknown for ten patients. The median number of hypermetabolic lesions identified for each patient was 6 (range 1 to 39). The median MTV was 93 cc (range 13 to 1,306 cc). Nineteen patients received RT near the time of immunotherapy initiation, with target volumes encompassing <1% to >99% of the total MTV (median 25%). The median follow-up duration for living patients is 11 months. MTV was a significant predictor of OS (HR 1.30 per doubling, 95% CI 1.00 to 1.69, p=0.047). The actuarial 12-month OS rate was 87% for patients with MTV less than or equal to 173 cc, compared to 74% for other patients. Lesions was a significant predictor of both OS (HR 1.05 per lesion, 95% CI 1.00 to 1.11, p=0.045)and PFS (HR 1.05 per lesion, 95% CI 1.01 to 1.08, p=0.005). Among patients who received RT near the time of diagnosis, RT targeting less than 25% of the total MTV was associated with inferior OS (12-monthOS 36% v. 100%, logrank p=0.011).


PET-based measures of disease burden (MTV) and multifocality (nLesions) may be important prognostic
factors for advanced NSCLC patients treated with first-line immunotherapy. Disease ‘debulking’ with RT
could be explored as an adjunct to immunotherapy. Validation of our findings with larger datasets is

FDG-PET Metrics in Advanced Non-Small Cell Lung Cancer (NSCLC): A Modern Review and

AC Berkowitz, B Halmos, H Cheng, C Huntzinger, & N Ohri Albert Einstein College of Medicine and Montefiore Medical Center, Bronx, NY

To provide a systematic review and meta-analysis of published literature characterizing the prognostic value of pre-treatment, volume-based FDG-PET metrics in patients with advanced NSCLC.


We conducted systematic PubMed searches to identify studies describing the prognostic value of volume-
based PET metrics (total metabolic tumor volume [MTV] and/or total lesion glycolysis [TLG]) obtained
prior to initiation of first-line systemic therapy for advanced (at least 50% Stage IV) NSCLC. The clinical
endpoints examined were progression-free survival (PFS) and overall survival (OS). Hazard ratios for PFS and OS were extracted directly from the original reports when available or estimated from survival curves using customized scripts. Inverse variance meta-analyses were performed to assess associations between PET metrics and clinical outcomes.


Our electronic searches yielded 416 records. We identified 58 abstracts to assess further for eligibility and
found 14 full-text articles that met eligibility criteria. Upon investigation of similar articles, we identified
another 6 to include for analysis. One article was subsequently found to have been retracted and was
excluded, along with 7 others which had either duplicate or insufficient data. We ultimately analyzed 12
articles, including 984 patients. Percentage of patients with Stage IV disease ranged from 67% to 100%.
Patients from at least 9 studies received chemotherapy and patients from at least 4 studies were treated with targeted therapy. No studies examining PET metrics for patients treated with first-line immunotherapy were identified. The median cutoff used to define high MTV across studies was 95 cc. Fixed-effects models demonstrated that high MTV is significantly associated with inferior PFS (HR=3.22, 95% CI 2.32 to 4.48, p<0.001) and inferior OS (HR=2.71, 95% CI 2.16 to 3.41, p<0.001). The median cutoff used to define high TLG across studies was 443 cc. Fixed-effects models demonstrated that high TLG is significantly associated with inferior PFS (HR=2.37, 95% CI 1.92 to 2.91, p<0.001) and inferior OS (HR=2.13, 95% CI1.80 to 2.51, p<0.001).


Baseline PET metrics (MTV, TLG) are powerful prognostic factors for advanced NSCLC patients who are
treated with chemotherapy or targeted therapy. Future studies should examine the prognostic value of PETmetrics for patients who receive first-line immunotherapy. Improved understanding of prognostic factors in this setting will be particularly important as we explore the role of radical local therapy for patients with limited disease burden.

Back To Top

ASTRO 20: RefleXion Presents New Research on Biology-Guided Radiotherapy for Metastatic Cancer.  Appl Radiol. 

By News Release| November 01, 2020

About the Author

News Release

News Release

Copyright © Anderson Publishing 2020