Physical versus biological dose for normal tissue complication probability in hypofractionated radiation therapy of unresectable liver cancer

Abstract

Purpose/ObjectiveAs the whole liver tolerance to radiation is low, escalated dose radiation therapy (RT) can only be delivered safely if a sufficient volume of liver is avoided. The risk of radiation induced liver disease (RILD) can be estimated using a normal tissue complication probability (NTCP) model. Our institution uses an NTCP model in a 6 fraction dose escalation study in which each patient has individualized prescription doses to maintain iso-toxicity. The starting dose level in this study is the prescription dose leading to 5% NTCP (or a maximum of 8 Gy per fraction). Currently, NTCP is estimated using physical dose-volume histograms (DVH). However, there is variation in the dose throughout the liver, and if corrections are made for dose per fraction to create a biologically normalized DVH, the estimated NTCP may change.The objectives of this study are to: 1) determine differences in NTCP based on physical and biologically normalized dose (BND), and 2) determine differences in iso-toxic prescription doses using physical dose and BND for NTCP analysis.Materials/MethodsFourteen patients with liver cancer (6 primary, 8 metastatic) planned with conformal RT were investigated. The effective liver volume ranged from 38–68%. An α/β of 2.5 Gy for normal liver was assumed as a conservative estimate for late responding tissue. NTCP was calculated using Lyman-Burman-Kutcher NTCP model parameters for RILD obtained from patients treated at 1.5 Gy per fraction (n = 0.97, m = 0.12, TD50(metastases) = 45.8 Gy, TD50(primary cancer) = 39.8 Gy)1Dawson L.A. et al.Int J Radiat Oncol Biol Phys. 2002; 53: 810-821Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar, with the TD50 value normalized to the individual patient prescription dose per fraction. Dose per fraction varied from 4.8–6.9 Gy, resulting in TD50 values between 16.9–25.1 Gy (as prescription dose varied to maintain 5% NTCP). For the BND calculation, a linear-quadratic transformation of the entire physical DVH to the prescription dose per fraction was performed. Differences in NTCP between the two methods were compared, and differences in dose prescriptions for iso-toxic levels were obtained.ResultsNTCP calculation with BND resulted in lower NTCP values than using physical dose for all patients. The mean reduction in NTCP from physical dose to BND was 3.3% (range: 1.3%–4.5%), resulting in BND NTCPs ranging from 0.1%–2.5%.Since BND is associated with decreased NTCP, the prescription doses associated with an NTCP of 5% were increased using BND compared to physical dose. Differences in the total prescription dose ranged from 0.6–4.8 Gy (in 6 fractions), as shown in figure 1. The mean prescription dose based on physical dose was 31.2 Gy, while the mean BND based prescription dose was 33.7 Gy, resulting in a mean increase in biologically effective tumour dose of 11% (assuming α/β = 10 Gy).ConclusionsBND results in a lower estimated NTCP than physical dose. For an iso-toxic prescription dose, the lower NTCP calculated with BND allows a higher prescription dose. We plan to use BND in the NTCP calculation for future patients. Future research involves validation of this NTCP model for hypofractionated RT and determination of α/β for RILD. Purpose/ObjectiveAs the whole liver tolerance to radiation is low, escalated dose radiation therapy (RT) can only be delivered safely if a sufficient volume of liver is avoided. The risk of radiation induced liver disease (RILD) can be estimated using a normal tissue complication probability (NTCP) model. Our institution uses an NTCP model in a 6 fraction dose escalation study in which each patient has individualized prescription doses to maintain iso-toxicity. The starting dose level in this study is the prescription dose leading to 5% NTCP (or a maximum of 8 Gy per fraction). Currently, NTCP is estimated using physical dose-volume histograms (DVH). However, there is variation in the dose throughout the liver, and if corrections are made for dose per fraction to create a biologically normalized DVH, the estimated NTCP may change.The objectives of this study are to: 1) determine differences in NTCP based on physical and biologically normalized dose (BND), and 2) determine differences in iso-toxic prescription doses using physical dose and BND for NTCP analysis. As the whole liver tolerance to radiation is low, escalated dose radiation therapy (RT) can only be delivered safely if a sufficient volume of liver is avoided. The risk of radiation induced liver disease (RILD) can be estimated using a normal tissue complication probability (NTCP) model. Our institution uses an NTCP model in a 6 fraction dose escalation study in which each patient has individualized prescription doses to maintain iso-toxicity. The starting dose level in this study is the prescription dose leading to 5% NTCP (or a maximum of 8 Gy per fraction). Currently, NTCP is estimated using physical dose-volume histograms (DVH). However, there is variation in the dose throughout the liver, and if corrections are made for dose per fraction to create a biologically normalized DVH, the estimated NTCP may change. The objectives of this study are to: 1) determine differences in NTCP based on physical and biologically normalized dose (BND), and 2) determine differences in iso-toxic prescription doses using physical dose and BND for NTCP analysis. Materials/MethodsFourteen patients with liver cancer (6 primary, 8 metastatic) planned with conformal RT were investigated. The effective liver volume ranged from 38–68%. An α/β of 2.5 Gy for normal liver was assumed as a conservative estimate for late responding tissue. NTCP was calculated using Lyman-Burman-Kutcher NTCP model parameters for RILD obtained from patients treated at 1.5 Gy per fraction (n = 0.97, m = 0.12, TD50(metastases) = 45.8 Gy, TD50(primary cancer) = 39.8 Gy)1Dawson L.A. et al.Int J Radiat Oncol Biol Phys. 2002; 53: 810-821Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar, with the TD50 value normalized to the individual patient prescription dose per fraction. Dose per fraction varied from 4.8–6.9 Gy, resulting in TD50 values between 16.9–25.1 Gy (as prescription dose varied to maintain 5% NTCP). For the BND calculation, a linear-quadratic transformation of the entire physical DVH to the prescription dose per fraction was performed. Differences in NTCP between the two methods were compared, and differences in dose prescriptions for iso-toxic levels were obtained. Fourteen patients with liver cancer (6 primary, 8 metastatic) planned with conformal RT were investigated. The effective liver volume ranged from 38–68%. An α/β of 2.5 Gy for normal liver was assumed as a conservative estimate for late responding tissue. NTCP was calculated using Lyman-Burman-Kutcher NTCP model parameters for RILD obtained from patients treated at 1.5 Gy per fraction (n = 0.97, m = 0.12, TD50(metastases) = 45.8 Gy, TD50(primary cancer) = 39.8 Gy)1Dawson L.A. et al.Int J Radiat Oncol Biol Phys. 2002; 53: 810-821Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar, with the TD50 value normalized to the individual patient prescription dose per fraction. Dose per fraction varied from 4.8–6.9 Gy, resulting in TD50 values between 16.9–25.1 Gy (as prescription dose varied to maintain 5% NTCP). For the BND calculation, a linear-quadratic transformation of the entire physical DVH to the prescription dose per fraction was performed. Differences in NTCP between the two methods were compared, and differences in dose prescriptions for iso-toxic levels were obtained. ResultsNTCP calculation with BND resulted in lower NTCP values than using physical dose for all patients. The mean reduction in NTCP from physical dose to BND was 3.3% (range: 1.3%–4.5%), resulting in BND NTCPs ranging from 0.1%–2.5%.Since BND is associated with decreased NTCP, the prescription doses associated with an NTCP of 5% were increased using BND compared to physical dose. Differences in the total prescription dose ranged from 0.6–4.8 Gy (in 6 fractions), as shown in figure 1. The mean prescription dose based on physical dose was 31.2 Gy, while the mean BND based prescription dose was 33.7 Gy, resulting in a mean increase in biologically effective tumour dose of 11% (assuming α/β = 10 Gy). NTCP calculation with BND resulted in lower NTCP values than using physical dose for all patients. The mean reduction in NTCP from physical dose to BND was 3.3% (range: 1.3%–4.5%), resulting in BND NTCPs ranging from 0.1%–2.5%. Since BND is associated with decreased NTCP, the prescription doses associated with an NTCP of 5% were increased using BND compared to physical dose. Differences in the total prescription dose ranged from 0.6–4.8 Gy (in 6 fractions), as shown in figure 1. The mean prescription dose based on physical dose was 31.2 Gy, while the mean BND based prescription dose was 33.7 Gy, resulting in a mean increase in biologically effective tumour dose of 11% (assuming α/β = 10 Gy). ConclusionsBND results in a lower estimated NTCP than physical dose. For an iso-toxic prescription dose, the lower NTCP calculated with BND allows a higher prescription dose. We plan to use BND in the NTCP calculation for future patients. Future research involves validation of this NTCP model for hypofractionated RT and determination of α/β for RILD. BND results in a lower estimated NTCP than physical dose. For an iso-toxic prescription dose, the lower NTCP calculated with BND allows a higher prescription dose. We plan to use BND in the NTCP calculation for future patients. Future research involves validation of this NTCP model for hypofractionated RT and determination of α/β for RILD.