1
|
Mo A, Velten C, Jiang JM, Tang J, Ohri N, Kalnicki S, Mirhaji P, Nemoto K, Aasman B, Garg M, Guha C, Brodin NP, Kabarriti R. Improving Adjuvant Liver-Directed Treatment Recommendations for Unresectable Hepatocellular Carcinoma: An Artificial Intelligence-Based Decision-Making Tool. JCO Clin Cancer Inform 2022; 6:e2200024. [PMID: 35671414 PMCID: PMC9225499 DOI: 10.1200/cci.22.00024] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022] Open
Abstract
PURPOSE Liver-directed therapy after transarterial chemoembolization (TACE) can lead to improvement in survival for selected patients with unresectable hepatocellular carcinoma (HCC). However, there is uncertainty in the appropriate application and modality of therapy in current clinical practice guidelines. The aim of this study was to develop a proof-of-concept, machine learning (ML) model for treatment recommendation in patients previously treated with TACE and select patients who might benefit from additional treatment with combination stereotactic body radiotherapy (SBRT) or radiofrequency ablation (RFA). METHODS This retrospective observational study was based on data from an urban, academic hospital system selecting for patients diagnosed with stage I-III HCC from January 1, 2008, to December 31, 2018, treated with TACE, followed by adjuvant RFA, SBRT, or no additional liver-directed modality. A feedforward, ML ensemble model provided a treatment recommendation on the basis of pairwise assessments evaluating each potential treatment option and estimated benefit in survival. RESULTS Two hundred thirty-seven patients met inclusion criteria, of whom 54 (23%) and 49 (21%) received combination of TACE and SBRT or TACE and RFA, respectively. The ML model suggested a different consolidative modality in 32.7% of cases among patients who had previously received combination treatment. Patients treated in concordance with model recommendations had significant improvement in progression-free survival (hazard ratio 0.5; P = .007). The most important features for model prediction were cause of cirrhosis, stage of disease, and albumin-bilirubin grade (a measure of liver function). CONCLUSION In this proof-of-concept study, an ensemble ML model was able to provide treatment recommendations for HCC who had undergone prior TACE. Additional treatment in line with model recommendations was associated with significant improvement in progression-free survival, suggesting a potential benefit for ML-guided medical decision making.
Collapse
Affiliation(s)
- Allen Mo
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Christian Velten
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY.,Institute for Onco-Physics, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Julie M Jiang
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Justin Tang
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Nitin Ohri
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Shalom Kalnicki
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Parsa Mirhaji
- Department of Systems & Computational Biology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY.,Center for Health Data Innovation, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Kei Nemoto
- Center for Health Data Innovation, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Boudewijn Aasman
- Center for Health Data Innovation, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Madhur Garg
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Chandan Guha
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY.,Institute for Onco-Physics, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - N Patrik Brodin
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY.,Institute for Onco-Physics, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| | - Rafi Kabarriti
- Department of Radiation Oncology, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY
| |
Collapse
|
2
|
Kim N, Kim HJ, Won JY, Kim DY, Han KH, Jung I, Seong J. Retrospective analysis of stereotactic body radiation therapy efficacy over radiofrequency ablation for hepatocellular carcinoma. Radiother Oncol 2018; 131:81-87. [PMID: 30773192 DOI: 10.1016/j.radonc.2018.12.013] [Citation(s) in RCA: 56] [Impact Index Per Article: 9.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/27/2018] [Revised: 12/09/2018] [Accepted: 12/12/2018] [Indexed: 12/12/2022]
Abstract
BACKGROUND AND PURPOSE To evaluate the efficacy of stereotactic body radiotherapy (SBRT) and radiofrequency ablation (RFA) for hepatocellular carcinoma (HCC). METHODS AND MATERIALS Patients treated for HCC between 2012 and 2016 were reviewed. Among these, 668 patients who underwent RFA of 736 tumors and 105 patients who underwent SBRT of 114 tumors were included. Using propensity score matching (PSM) to adjust for clinical factors, 95 tumors were selected from each treatment arm. Freedom from local progression (the primary endpoint, FFLP) was compared before and after adjustment with PSM. RESULTS At baseline, SBRT-treated tumors were more advanced, larger (median, 2.4 vs. 1.6 cm), and more frequently located in the subphrenic region than RFA-treated tumors (P < .001). The median follow-up was 21.5 (interquartile range, 11.2-36.7) months. Before PSM, the 2-year FFLP rates were 76.3% for the SBRT group and 70.2% for the RFA groups, respectively. After PSM, the 2-year FFLP rates were 74.9% for the SBRT group and 64.9% for the RFA group, respectively. The local control rates were not significantly different. The Cox proportional hazards model revealed the treatment modality as an independent predictor of local recurrence favoring SBRT in the entire cohort and in the PSM model. Elevated tumor markers, tumor location (subphrenic region), and tumor size (>2.0 cm) were also independent predictors of local progression. CONCLUSION SBRT appears to be an effective alternative treatment for HCC when RFA is not feasible due to tumor location or size.
Collapse
Affiliation(s)
- Nalee Kim
- Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Hyun Ju Kim
- Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Jong Yun Won
- Department of Radiology, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Do Young Kim
- Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Kwang-Hyub Han
- Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Inkyung Jung
- Department of Biostatistics & Medical Informatics, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Jinsil Seong
- Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea.
| |
Collapse
|