In an interview with Oncology Learning Network, Cheng-En Hsieh, MD, answers a few questions about a multi-center study he led aiming to identify predictors of radiation-induced liver disease (RILD) in patients with hepatocellular carcinoma (HCC) treated with proton beam therapy (Int J Radiat Oncol Biol Phys. 2019;105[1]:73-86).

Dr Hsieh is a Visiting Scientist at the University of Texas MD Anderson Cancer Center, where he is also pursuing a PhD in the UTHealth Graduate School of Biomedical Sciences. Previously, Dr Hsieh was an attending physician in the Department of Radiation Oncology at Chang Gung Memorial Hospital in Taiwan and specialized in gastrointestinal and head and neck oncology.

Dr Hsieh’s current research projects focus on advancing the therapeutic efficacy of radiation therapy, reducing radiation toxicity, as well as identifying clinical, biological and imaging prognosticators for gastrointestinal and head and neck cancers.

What existing data led you and your co-investigators to conduct this research?

The role of proton therapy in the definitive management of localized liver cancer has been expanding rapidly, particularly to treat patients with larger tumors or major vessel involvement. Nonetheless, liver tissue is intrinsically vulnerable to radiation, and RILD could lead to hepatic failure, especially in cirrhotic livers.

Establishing optimal dose-volumetric indices to prevent RILD is urgently needed for proper treatment planning with proton therapy. Due to inherent differences in the dose distribution characteristics of protons versus photons (x-rays), we postulated that using the same metrics used for evaluation of photon-based plans may not be applicable to the evaluation of proton-based plans. As a corollary, clinical practice may need to be adapted as we continue to treat greater numbers of liver cancer patients with protons.

Accordingly, we conducted this multi-institutional study to identify clinically useful dose-volumetric predictors for RILD prevention in patients with HCC treated with proton therapy.

Please briefly describe your study and its findings. Were any of the outcomes particularly surprising?

This multi-institutional study included 136 patients with HCC who were treated with high-dose hypofractionated proton therapy at Chang Gung Memorial Hospital in Taiwan (n = 102) or at MD Anderson Cancer Center in Houston (n = 34).

Our results show that radiation doses higher than 1 GyE can produce detrimental effects on the liver, and that preserving sufficient unirradiated liver volume is imperative to prevent RILD with this treatment approach. We identified the ratio of unirradiated liver volume to standard liver volume (ULV/SLV) as the most significant dose-volumetric predictor of RILD. We observed a consistent “volume-response” relationship between ULV/SLV and RILD in both eastern and western cohorts, where lower ULV/SLV correlated strongly with higher RILD risk.

Our data also show that the risk of RILD was minimal in CP class-A (CP-A) and CP class-B (CP-B) patients with ≥50% and ≥60% of ULV/SLV, respectively. On the contrary, substantial hepatic toxicities were recorded in CP-A and CP-B patients whose ULV/SLV were <30% and <40%, respectively.

With conventional photon radiation therapy, the risk of RILD is primarily determined by the mean liver dose (MLD). To stay within the MLD constraints, de-escalation of the target dose is frequently mandated for larger tumors, resulting in a considerably lower probability of tumor control.

With proton therapy, however, we found that the volume of liver untouched by radiation was more important than the dose of radiation delivered for RILD prevention. Our results indicate that if sufficient ULV/SLV is preserved, ablative doses can be delivered safely with minimal risk of treatment-related liver disease regardless of MLD or target dose.

What are the possible real-world applications of these findings in clinical practice?

The dose-volumetric indices used in our study are readily accessible in currently available treatment planning systems. Our results could help radiation oncologists weigh the risks and benefits of treatment for personalized decision-making in the clinic.

In light of the reality that many of the liver cancer patients lack better treatment alternatives due to the presence of bulky tumors, vascular involvement or severely compromised liver function, meticulous refinement of proton therapy treatment planning plays a crucial role in affording them an ablative treatment option with potentially promising tumor control outcomes. For these vulnerable patients, our findings suggest adopting only 2-3 proton beams with narrow beam angles in the design of proton treatment plans to maximize their ULV/SLV.

Furthermore, for image-guided liver-directed therapy, stringent respiratory motion management such as abdominal compression, breath-hold, and respiratory gating or tracking should be considered to minimize irradiation of uninvolved liver tissue.

Do you and your co-investigators intend to expand upon this research?

Ji-Hong Hong, MD, PhD, Chang Gung Memorial Hospital, Taiwan, Sunil Krishnan, MBBS, MD, Mayo Clinic, Jacksonville, Florida, and I are currently conducting a larger multi-institutional study to devise a nomogram model combining dose-volumetric parameters and clinical characteristics to predict radiation-induced liver morbidity in HCC patients treated with proton therapy.

In addition, we are also comparing outcomes for patients treated with proton therapy to outcomes for those treated with conventional x-ray radiation therapy to identify which subgroups of patients could benefit the most from proton therapy.

Disclosure

This work was supported by funding from the Chang Gung Memorial Hospital grant CIRPG3D0141 and CIRPG3F0021, Taiwan Ministry of Science and Technology grant MOST 106-2627-M-182A-001, the MD Anderson Cancer Center support grant P30 CA16672 and the John E. and Dorothy J. Harris Endowed Professorship (to Dr. Sunil Krishnan).