Circulating Tumor DNA for Early Risk Stratification of Oligometastatic Lung Cancer

PURPOSE/OBJECTIVE(S)

Some patients with oligometastatic disease experience prolonged progression-free survival when treated with local consolidative radiotherapy (RT). Precisely identifying these patients remains challenging, however. We hypothesized that pre-RT liquid biopsy circulating tumor DNA (ctDNA) analysis could risk-stratify oligometastatic non-small-cell lung cancer (NSCLC) patients and enable earlier personalized selection for consolidative RT.

MATERIALS/METHODS

A real-world multi-institutional cohort of 1,487 patients who were diagnosed with oligometastatic NSCLC was analyzed. Each patient underwent liquid biopsy ctDNA analysis using the Tempus xF assay (v2) at least once, for a total of 1,880 ctDNA assays. 20% of the cohort (n = 309) underwent RT after liquid biopsy was obtained and oligometastatic NSCLC was diagnosed by the treating physician. Each patient in the sub-cohort of 309 patients had oligometastatic disease, defined as metastatic disease present in 1-5 organ systems. Outcomes for overall survival (OS) and progression-free survival (PFS) were defined with respect to the initiation time of RT to minimize the risk of guarantee-time bias. ctDNA results were analyzed for variants using VarDict and characterized as pathogenic or likely pathogenic following ACMG/AMP guidelines for variant classification, as determined by SnpEff. Variants considered benign, likely benign, or having conflicting evidence were excluded from consideration.

RESULTS

Overall survival was significantly worse in oligometastatic NSCLC patients with detectable ctDNA pre-RT, as compared to those without detectable ctDNA pre-RT, with a median OS of 16.8 months versus 25 months (p = 0.030, HR = 1.65, CI = 1.05-2.61). Similar findings were also observed for PFS, which was worse in patents with detectable ctDNA pre-RT, with a median PFS of 5.4 months versus 8.8 months (p = 0.004, HR = 1.57, CI = 1.15-2.13). ctDNA variant allele frequency (VAF) levels demonstrated significant risk correlations, with the maximum pre-RT ctDNA VAF associated with increased risk of both disease progression (p = 0.0084) and death (p = 0.0073). These findings were corroborated by multivariate Cox proportional hazards modeling for PFS (p = 0.02, PFS HR = 4.69, CI = 1.42-13.30) and OS (p = 0.004, HR = 5.66, CI = 1.64-16.85). Notably, multivariate Cox modeling did not show significant impacts of other clinical parameters, including gender, age at diagnosis, smoking status, and squamous histology. Additionally, the ctDNA mutational burden (the number of detectable pathogenic or likely pathogenic variants) was significantly associated with risk for both PFS (p = 0.003, HR = 1.16, CI = 1.06-1.26) and OS (p = 0.003, HR = 1.15, CI = 1.04-1.25) in a multivariate Cox regression model.

CONCLUSION

These data suggest that liquid biopsy ctDNA detection represents a powerful pre-RT biomarker to risk-stratify oligometastatic NSCLC patients and potentially enable personalized decision-making for local consolidative RT.

A Pilot Study of Simulation-Free Hippocampal-Avoidance Whole Brain Radiotherapy Using Diagnostic MR-Based and Online Adaptive Planning

PURPOSE/OBJECTIVE(S)

We aimed to demonstrate the clinical feasibility and safety of a simulation-free hippocampal avoidance whole brain radiotherapy (HA-WBRT) workflow in a Phase I clinical trial (NCT05096286). Feasibility was defined as successful completion of the simulation-free HA-WBRT workflow through treatment delivery in at last 70% of treated patients.

MATERIALS/METHODS

Ten candidates for HA-WBRT were enrolled for treatment on a ring gantry CT-guided Linac with online adaptive capabilities. Structures were contoured on the diagnostic brain MRI, which was then registered to a separate head computed tomography (CT) of similar head shape, obtained from an atlas-based database. A HA-WBRT "pre-plan" was generated using the atlas-based CT (AB-CT) and the NRG-CC001 constraints. At first fraction, the AB-CT was used as the primary dataset and deformed to the patient's cone-beam CT (CBCT) for dose calculation. The brain, ventricles, and brainstem contours were matched through rigid translation and rotation to the corresponding anatomy on the CBCT to aid in alignment, given the differences in rotational head positioning from diagnostic MRI to CBCT setup. Lastly, the lens, optic nerves, and brain contours were manually edited based on CBCT visualization. Plans were then optimized, and the adaptive plan was chosen for treatment if the plan met all objectives. Workflow tasks were timed. In addition, conventional plans using patients' sim CTs were created for each patient for the purpose of prospective dosimetric comparison. The dosimetric parameters were compared for each patient between the delivered sim-free plan and the conventional sim CT plan using the sign test via statistical software, with p<.05 indicating significance.

RESULTS

Median time from approved sim order to first fraction was 4 days (range: 2-7); median time in room (door-to-door) was 49 minutes (range: 35-70). All patients successfully completed all ten fractions and 90% of the simulation-free radiation plans met all NRG-CC001 constraints. For one patient, the sim-free plan at fraction one failed the planning target volume (PTV) coverage objective (coverage of 89%); this was deemed acceptable for delivery by the treating radiation oncologist. An offline replan was then performed to meet NRG-CC001 constraints and used for the subsequent nine fractions. There was no clinically meaningful difference in dosimetric constraints between the sim-free plan (calculated on AB-CT) and conventional CT sim plan. Statistically, the sim-free plans provided improved PTV coverage to higher doses compared to the conventional plans (Table). At a median follow-up of 43 days (range: 9 -280), the intracranial progression-free survival rate was 90%.

CONCLUSION

Simulation-free HA-WBRT is feasible, results in plans that are dosimetrically comparable to conventional CT sim workflows and succeeds in decreasing time to initiation of HA-WBRT by at least 50%. Further studies with a larger cohort are warranted to optimize the workflow.

The Impact of Intra-Fraction Bowel Motion on Luminal Gastrointestinal Organ at Risk Dosimetry When Using Stereotactic Adaptive Radiotherapy for Abdominal Malignancies

PURPOSE/OBJECTIVE(S)

Daily online adaptive radiotherapy (ART) provides inter-fraction motion management of the luminal gastrointestinal (GI) structures when delivering abdominal SBRT. One potential drawback of ART is the time-consuming process, and intra-fraction GI changes from completion of the ART process to the end of treatment delivery have not been thoroughly evaluated. We explored intra-fraction bowel motion for patients receiving abdominal stereotactic adaptive radiotherapy (sART) MATERIALS/METHODS: Six patients with abdominal malignancies treated with CT-guided sART on a prospective feasibility trial had additional CBCT's acquired post-treatment (pTx-CBCT). All patients were prescribed to 50 Gy/5 fractions (fx), and the constraint for all GI OARs was V33≤0.5 cc. Time from initial CBCT (I-CBCT) used for adaptive planning to pTx-CBCT was collected. The luminal GI OAR (stomach (S), duodenum (D), small bowel (SB), and large bowel (LB)) were retrospectively contoured on pTx-CBCT. The OAR doses were compared between the I-CBCT and pTx-CBCT. The adaptive plan (PA) and initial plan (PI) doses were overlayed on the pTx-CBCT contours. The PA pTx-CBCT OAR doses were then compared to the PI pTx-CBCT OAR doses. A Boolean OAR structure of all GI OARs was evaluated to remove potential differences in structure definitions between providers. The T-test was used to compare differences in instances of D0.5cc ≥ 33 and 50 Gy. Patient charts were reviewed for grades (G) ≥ 3 toxicity.

RESULTS

Thirty fractions (fx) of sART were delivered and pTx-CBCT were acquired in 26 fx. Mean time from I-CBCT to pTx-CBCT acquisition was 66 min (38-98 min). On average at 0.5 cc the PA overdosed the S by 1.74 Gy based on pTx-CBCT anatomy compared to 2.35 Gy by the PI, the D by 0.47 Gy (PA) vs .84 Gy (PI), the SB by 1.14 Gy (PA) vs 1.43 Gy (PI), and the LB by 0.13 Gy (PA) vs 0.60 Gy (PI). The dose to the Boolean OAR structure was on average 2.51 Gy/fx higher than expected when overlaying the PA on the pTx-CBCT compared to 3.38 Gy/fx higher when overlaying the PI on the pTx-CBCT. There was no significant difference in the instance of the PA exceeding D0.5 cc ≥33 Gy vs the PI (p = 0.083), but the PA significantly reduced the instances of D0.5cc≥50 Gy (p = 0.001) compared to the PI. No patient experienced G≥3 toxicity at a median follow-up of 8 months (3-12).

CONCLUSION

These data demonstrate sART led to a significant decrease in dose to GI OARs, particularly for prescription dose or greater, even after accounting for intra-fractional bowel motion. While both the PI and the PA violated the V33 luminal GI OAR constraint in approximately ½ of pTx-CBCTs, the fraction of OARs receiving at least 50 Gy was significantly higher when overlaying the PI compared to the PA. While no G3 toxicities were reported in this small cohort, further studies are needed to characterize if the increased dose to GI OARs over the expected dose is clinically significant.

Patient Selection is Critical When Evaluating Candidates for 5-Fraction Stereotactic MRI-Guided Adaptive Radiotherapy (SMART) for Locally Advanced Pancreatic Cancer

PURPOSE/OBJECTIVE(S)

Stereotactic magnetic resonance image-guided adaptive radiotherapy (SMART) allows for the safe delivery of biologically effective doses (BED10 ∼ 100 Gy) to patients with pancreatic cancer who otherwise have limited therapeutic avenues. In this study, we analyze long-term outcomes in the largest cohort of patients with borderline-resectable (BR), locally advanced (LA), or medically inoperable (MI) pancreatic cancer treated with a 5-fraction SMART technique.

MATERIALS/METHODS

A single institution analysis of patients with BR, LA, or MI pancreatic cancer treated with SMART between 2015 and 2021 was performed. Patients with locally recurrent disease, non-adenocarcinoma histology, de-novo metastatic disease, or who did not receive induction chemotherapy were excluded. Baseline and treatment characteristics were collected. Local control (LC), progression free survival (PFS), and overall survival (OS) were calculated using the Kaplan-Meier method, and factors associated with outcomes were evaluated using Cox regression analyses. Oncologic outcomes measured from time of initial diagnosis; local control defined as freedom from local progression.

RESULTS

A total of 129 patients were reviewed. Median age at diagnosis was 67 years. The majority were male (60%) and White (84%). 23% of patients had an ECOG 2. Most patients had LA disease (66%), followed by BR (20%) and MI (14%). Median follow up was 17.0 months (6-61 months). Median LC was not reached, and LC was 81%, 54%, and 48% at 1, 2, and 3-years, respectively. Median PFS was 12.9 months [95% confidence interval 11.8-14.71] with 1, 2, and 3-year PFS of 60%, 20%, and 7%, respectively. Median OS was 17.7 months [15.7-19.7] with 1, 2, and 3-year OS of 78%, 28%, 11%, respectively. On univariate analysis, increased duration of induction chemotherapy had a statistically significant impact on PFS and OS. Those receiving equal to or greater than 4 months of induction chemotherapy had a median OS of 19.2 months [17.2-21.2] as compared to 13.6 [11.9-15.3] for those with less than 4 months.

CONCLUSION

In this patient population which included a large portion of patients with ECOG of 2 or greater and those deemed MI, a 5-fraction SMART regimen yielded durable long-term LC. The impact of increasing duration of induction chemotherapy underlies the importance of patient selection and improved understanding of tumor-specific biology when selecting patient's with locally advanced pancreatic cancer for aggressive local therapy.