A Prospective Study of Bevacizumab and Neoadjuvant Concurrent Chemoradiation in Locally Advanced Esophageal Squamous Cell Carcinoma: Paradoxical Increase in Circulating Vascular Endothelial Growth Factor-A and Effect on Outcome

PURPOSE/OBJECTIVE(S)

In the prior prospective biomarker study, high serum vascular endothelial growth factor-A (VEGF-A) was associated with a poor prognosis. We conducted a prospective phase II trial of adding Bevacizumab, an anti-VEGF-A monoclonal antibody, to neoadjuvant concurrent chemoradiation (neoCCRT) for patients with locally advanced esophageal squamous cell carcinoma (LA-ESCC). This prospective biomarker study aims to evaluate the expressions of angiogenesis-associated circulating biomarkers before and after neoCCRT and compare clinical outcomes for patients receiving platinum/5-fluorouracil (PF) with or without Bevacizumab.

MATERIALS/METHODS

Patients with biopsy-proven resectable non-T4 LA-ESCC were enrolled for the prospective phase II trial investigating PF-neoCCRT plus Bevacizumab (BPF group). A parallel patient cohort enrolled in a prospective biomarker study receiving PF-neoCCRT was included in the present analysis as the control group (PF group). Radiotherapy was delivered with 40 Gy in 20 fractions. All patients had restaging workups after enoCCRT and underwent radical esophagectomy if the disease remained resectable. Serums were collected before and after neoCCRT. The serum concentrations of angiogenesis-associated biomarkers were determined by the multiplex enzyme-linked immunosorbent assay. Survival analyses were performed by the Kaplan-Meier method. The t-test and log-rank test were used to compare differences in biomarker expression and survival between groups.

RESULTS

From 2016 to 2019, 43 patients (BPF/PF group: 21/22) were enrolled in the study. Twenty patients in each group had serum samples available for biomarker analysis. 15 out of 21 patients in the BPF group and 20 out of 22 patients in the PF group underwent radical esophagectomy. Six patients in the BPF group and nine patients in the PF group achieved pathological complete responses. The median overall survival for the BPF and PF group was 20.8 months and not-reached, respectively (hazard ratio = 1.33, long rank p = 0.58). In the BPF group, the serum VEGF-A level was significantly increased from an average value of 446 pg/mL to 723 pg/mL after neoCCRT (p = 0.037), while its level was decreased from 815 ng/mL to 380 pg/mL in the PF group (p = 0.104). In addition, the expression value of circulating Angiopoietin-1 was not changed in the BPF group (before neoCCRT, mean value = 828 pg/mL; after neoCCRT, mean value 762 pg/mL, p = 0.67). In contrast, serum Angiopoietin-1 level was reduced from an average value of 659 pg/mL before neoCCRT to 271 pg/mL after neoCCRT (p = 0.002) in the PF group.

CONCLUSION

The addition of Bevacizumab to PF-neoCCRT did not improve pathological or survival outcomes in patients with resectable LA-ESCC. Adding a single dose of Bevacizumab paradoxically increases circulating VEGF-A while maintaining the Angiopoietin-1 serum level after neoCCRT. Further investigation by using additional VEGF-A inhibition may be required to achieve sustained angiogenesis blocked for tumor control.

Automated Detection, Segmentation, and Tracking of Brain Metastases in Repeated Courses of Stereotactic Radiosurgery Using Integrated Artificial Intelligence

PURPOSE/OBJECTIVE(S)

Salvage stereotactic radiosurgery (SRS) for distant brain metastases has been demonstrated as a safe and effective approach for intracranial recurrences after initial SRS. However, accurate tumor detection and segmentation among responding tumors within the irradiated parenchyma can be challenging. The requirement for the registration and reference to the previous course of SRS is very time-consuming and suffers significant inter and intra-reader variability. Artificial intelligence (AI)-assisted system has been proven to improve the accuracy and efficiency in the clinical flow of de-novo SRS. We hypothesize that an integrated AI system can facilitate an automated tumor contouring process for repeated SRS.

MATERIALS/METHODS

Three patients who underwent their third course of SRS to brain metastases were selected for the pioneering works. They have had two sessions of SRS with a mean lesion number of 4 and 3.7, respectively. VBrain, an FDA-approved brain tumor management AI platform, was used to co-registered serial MR scans and automatically identify, track, and contour brain metastases for each course of SRS. The AI also indicated new lesions and treated lesions for each course. Three radiation oncologists experienced in brain SRS contoured the gross tumor volumes (GTVs) of the third course of SRS in two reader modes (assisted then unassisted) with a memory washout period of one week between each section. The segmentation ground truth was established through consensus among the three experts. Lesion-wise sensitivity, contouring accuracy, and consuming time were compared between the two contouring modes.

RESULTS

In each patient, there were 15, 11, and 9 metastases, with a median diameter of 4.72 (95% CI: 4.05, 6.91) mm. The mean lesion-wise sensitivity was 96.96±2.47% with AI assistance and 76.90 ± 7.10% without assistance. There were two false-positive lesions in the assisted read, resulting in a low average false-positive rate of 0.67 per patient, while no false positive for the unassisted mode. AI assistance improved contouring accuracy. The median Dice similarity coefficient (DSC) was 0.71 (95% CI: 0.55, 0.87) for assisted contouring and 0.65 (95% CI: 0.46, 0.85) for unassisted contouring. We also use average Hausdorff distance (HD) to measure segmentation results. The mean HD was 0.72± 0.13 mm versus 0.73±0.08 mm for the two contouring modes (p = 0.02) Furthermore, the median contouring time per case was significantly shorter with AI assistance than without assistance (20.8 minutes vs. 29.8 minutes; p < 0.001), corresponding to a 43.2% time-saving.

CONCLUSION

Our results suggest that the integration of an AI-based system into repeated brain SRS can significantly improve the accuracy and efficiency of tumor detection and segmentation. This approach has the potential to streamline the treatment planning process for salvage SRS.

Whole Liver Radiotherapy for Multiple Liver Metastases in Thoracic Malignancies

PURPOSE/OBJECTIVE(S)

Liver metastases (LM) are common in advanced thoracic cancer patients with dismal oncological outcomes. Even in the modern era of novel systemic therapy, LM led to a 21% increased mortality risk compared with those without. Options for progressive multiple LM after systemic therapy are limited. Therefore, a different treatment modality is urgently needed to overcome such a predicament. Herein, we renovate the classical whole liver radiotherapy (WLRT) and evaluate its efficacy and safety in treating thoracic cancer patients with progressive multiple LM.

MATERIALS/METHODS

Patients with lung or thymic cancer who had multiple LM treated by WLRT between 2018 and 2022 were enrolled. Radiotherapy (RT) was delivered with a median dose of 24 Gy (range 8-25 Gy) in 8 fractions (range 1-16) at the discretion of treating physicians. Overall survival (OS) and cumulative incidence of intrahepatic progression were calculated from the completion of WLRT till death or progression by Kaplan-Meier and competing risk analyses, respectively.

RESULTS

Twenty-four patients were enrolled in this retrospective study. Eleven patients (46%) had lung adenocarcinoma, of which nine patients had oncogenic mutations. Six patients (25%) had small cell lung cancer (SCLC), four patients (17%) had thymic squamous cell carcinoma (SqCC), two patients (8%) had lung SqCC, and one patient (4%) had mixed histology. Eighteen patients (75%) were under systemic therapy treatment before the diagnosis of LM, and fifteen (63%) received LM biopsy. The median time from the diagnosis of LM to WLRT was 7.5 months (range, 0.5-33.9 months). Eleven patients (46%) had concurrent RT and systemic therapy. With a median follow-up of 3.1 months, the 3-, 6- and 12-month OS were 57%, 38%, and 15%, respectively. After adjusting death as a competing risk, the cumulative incidence of LM progression at 3, 6, and 12 months were 10%, 19%, and 44%, respectively. Within three months after the completion of RT, one patient (4%) had grade 5 radiation-induced liver disease (RILD), one patient (4%) had grade 4 abnormal liver function test (LFT), three patients (13%) had grade 3 abnormal LFT, and twelve patients (50%) had ≤ grade 2 abnormal LFT. The patient who had Gr.5 RILD expired 51 days after the completion of RT with 24 Gy in 8 fractions concurrently with topotecan. He had primary SCLC with viral hepatitis B. His LFT was around the normal upper limit before WLRT.

CONCLUSION

WLRT provided favorable intrahepatic control with acceptable radiation-related toxicities and could be considered a treatment option for patients with progressive LM under systemic therapy. Further investigation with a larger cohort is warranted to identify patients at a high risk of developing severe RILD.

Comparison of Hippocampal Volume Change after Whole-Brain Radiotherapy with and without Hippocampal Avoidance: A Post-Hoc Analysis of a Phase II Randomized Trial

PURPOSE/OBJECTIVE(S)

Hippocampal avoidance whole-brain radiotherapy (HA-WBRT) preserves verbal memory compared to conformal WBRT in the previously reported randomized phase II clinical trial. In the present study, we analyzed the changes in hippocampus volume between groups at longitudinal follow-up time points. We hypothesize better preservation of hippocampus volumes with time in the HA-WBRT group.

MATERIALS/METHODS

In this post-hoc analysis of a single institution, single-blind, randomized phase II trial, patients with brain metastases were randomized to receive either HA-WBRT or conformal WBRT with 30 Gy in 10 fractions. Patients received gadolinium contrast-enhanced MRI every three months or when clinically indicated after WBRT until progression. Only patients with at least one follow-up MRI were included. The MRI was used to assess baseline intracranial volume, hippocampal volume, the ratio of the total hippocampus to total intracranial volume (HT%), and the ratio of the left (HL%) or right (HR%) hippocampus to total intracranial volume. The volumes were automatically calculated by volBrain, an online MRI brain volumetry system. We used linear mixed models with fixed and random effects to compare the changes of repeated measures of the hippocampus volumes over time between the two groups.

RESULTS

Among 70 randomized patients, 51 (25 in the HA-WBRT group and 26 in the WBRT group) were included in the present analysis. There were no significant differences between the two groups in the baseline demographics, HT%, HR%, and HL%. The median follow-up time was 13.9 months (range 2.1-31.8 months), and more than 60% of patients had at least one year of follow-up. Linear mixed model analyses showed a more significant decrease of HT% over time in the WBRT arm (regression coefficient β = -0.002, p = 0.001) compared to the HA-WBRT arm (regression coefficient, β = -0.001, p = 0.254), with a significant group by time interaction (F-value = 4.27, p = 0.050). The HL% was stable over time in the HA-WBRT arm (β = 0.0001, p = 0.769) while it decreased in the WBRT arm (β = -0.001, p = 0.003), with a significant group by time interaction (F-value = 6.51, p = 0.015). In contrast, the HR% decreased over time in both groups (HA-WBRT: regression coefficient β = -0.001, p = 0.006; WBRT: regression coefficient β = -0.001, p < 0.001) without significant group by time interaction (F-value = 0.20, p = 0.662).

CONCLUSION

HA-WBRT better preserves total hippocampus volume with time compared to WBRT, with the main effect contributed by preserving the left hippocampus volume. Further research is warranted to establish the relationship between hippocampus volume perpetuation and cognitive function preservation. The future goal is to investigate whether left-side unilateral HA-WBRT provides similar neurocognitive function preservation compared to standard bilateral HA-WBRT.