An update on NTCP data Vladimir Semenenko, Ph.D. X. Allen Li, Ph.D. Department of Radiation Oncology Medical College of Wisconsin NCCAAPM 2006 Fall Meeting, October 3, 2006
Goal of Radiation Therapy Maximize probability of tumor eradication Minimize probability of injury to surrounding normal tissues
Goal of Radiation Therapy Probability of Outcome Tumor Control Probability (TCP) TCP ( NTCP) Normal Tissue Complication Probability (NTCP) Radiation Dose
Goal of Radiation Therapy Maximize probability of tumor eradication Deliver sufficient dose to tumor cells Minimize probability of injury to surrounding normal tissues Reduce normal tissue doses to acceptable levels
Tissue Tolerance Data Emami et al. IJROBP 2, 09-22, 99 Review of literature and personal experience of the authors 28 critical sites, TD 5/5 and TD 50/5 for one-third, two-thirds thirds and the whole organ Data fit to the Lyman model by Burman et al. (IJROBP 2, 23-35, 35, 99) n, m, TD 50
Lyman-Kutcher Kutcher-Burman (LKB) Model Lyman Radiat.. Res. 04, S3-9, 985; Kutcher and Burman IJROBP 6, 623-30, 989 t 2 x 2 NTCP = e dx 2π 3 parameters: n, m, TD 50 t D TD eff 50 / = mtd 50 D n n eff = vd i i i n = Deff = Dmean = vd i i i
Lyman Model in Clinical Practice Dose-escalation escalation study for intrahepatic cancers; NTCP = 0% (McGinn( et al. JCO 6, 2246-52, 998) 3D-CRT dose-escalation escalation studies for NSCLC; NTCP 25% (Rozenzweig( et al. Cancer 03, 28-27, 27, 2005) SBRT (3 9-5 Gy/fr); NTCP 20% 4 out of 7 patients had grade -2 2 RP (Song et al. AJCO 28, 59-6, 2005)
Project Goal To identify reliable estimates of Lyman model parameters for the purposes of biologically-based based treatment planning
Lung Reference n m TD 50 (Gy) Endpoint Fractionation Scheme Lung Burman et al. 99 0.87 0.8 24.5 Pneumonitis.8-2 Gy q.d. Martel et al. 994 0.87 0.8 28 SWOG grade RP.8-2 Gy q.d. Kwa et al. 998 0.30 30.5 SWOG grade 2 RP -2.7 Gy q.d.; normalized to 2 Gy/fr using α/β of 2.5 or 3 Gy Seppenwoolde et al. 2003 0.99 0.37 0.28 30.8 43 SWOG grade 2 RP SWOG grade 3 RP -2.7 Gy q.d.; normalized to 2 Gy/fr using α/β of 2.5 or 3 Gy Moiseenko et al. 2003.02 0.26 2.0 Symptomatic pneumonitis -2 Gy q.d.; normalized to 2 Gy/fr using α/β of 3 Gy 0.80 0.37 2.9 Radiographic and symptomatic pneumonitis
Lung Observed RP Rate 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0. Graham et al. 999 (Washington U) - RTOG grade >=2 Moiseenko et al. 2003 (Canada) - RTOG grade >=2 Maximum likelihood fit - no steroids n = m = 0.40 TD50 = 29.5 Gy Observed RP Rate 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0. Seppenwoolde et al. 2003 (Netherlands) - SWOG grade >=2 Willner et al. 2003 (Germany) - NCI CTC grade >=2 Kim et al. 2005 (Korea) - RTOG grade >=3 Yorke et al. 2005 (MSKCC) - RTOG grade >=3 Chang et al. 2006 (U of Florida) - NCI CTC grade >=2 Maximum likelihood fit - steroids and worse n = m = 0.38 TD50 = 28. Gy 0 0 0 0 20 30 Maximum 40likelihood fit - 50 no steroids 0 0 20 30 40 50 0.9 Mean Lung Dose (Gy) Maximum likelihood fit - steroids and worse Mean Lung Dose (Gy) 0.8 Observed RP Rate 0.7 0.6 0.5 0.4 0.3 0.2 0. 0 0 0 20 30 40 50 Mean Lung Dose (Gy)
Lung Observed RP Rate 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0. 0 Graham et al. 999 (Washington U) - RTOG grade >=2 Seppenwoolde et al. 2003 (Netherlands) - SWOG grade >=2 Moiseenko et al. 2003 (Canada) - RTOG grade >=2 Willner et al. 2003 (Germany) - NCI CTC grade >=2 Kim et al. 2005 (Korea) - RTOG grade >=3 Yorke et al. 2005 (MSKCC) - RTOG grade >=3 Chang et al. 2006 (U of Florida) - NCI CTC grade >=2 Maximum likelihood fit - all RP n = m = 0.39 TD50 = 28.6 Gy 0 0 20 30 40 50 Mean Lung Dose (Gy)
Parotid Gland Reference n m TD 50 (Gy) Endpoint Fractionation Scheme Parotid gland Burman et al. 99 0.70 0.8 46 Xerostomia.8-2 Gy q.d. Eisbruch et al. 999 0.8 28.4 RTOG grade 4 toxicity (stimulated.8-2 Gy q.d. salivary flow rate reduced to 25% pre- RT flow at 2 months) Schilstra and Meertens.3 0.26 38 SSF 25% at 3 weeks 2 Gy q.d. 200 Roesink et al. 200 Braam et al. 2005 0.54 0.46 0.45 0.37 0.33 0.37 0.53 3 35 39 34 40 42 46 SSF 25% at 6 weeks SSF 25% at 6 months SSF 25% at 2 months SSF 25% at 6 weeks SSF 25% at 6 months SSF 25% at 2 months SSF 25% at 5 years 2 Gy q.d. 2 Gy q.d.
Parotid Gland Observed Complication Rate 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0. 0 n = m = 0.54 TD50 = 29.9 Gy Eisbruch et al. 999 (U of Michigan) Roesink et al. 200 (Netherlands) Scrimger et al. 2004 (Canada) Maximum likelihood fit 0 20 40 60 80 Mean Dose (Gy)
Esophagus Reference n m TD 50 (Gy) Endpoint Fractionation Scheme Esophagus Burman et al. 99 0.06 0. 68 Clinical stricture/perforation.8-2 Gy q.d. Belderbos et al. 2005 0.69 0.36 47 RTOG grade 2 AE 2.25-2.75 Gy q.d.; normalized to 2 Gy/fr using α/β of 0 Gy Chapet et al. 2005 0.44 0.32 5 RTOG grade 2 AE 2. Gy q.d.; normalized to 2 Gy/fr using α/β of 0 Gy This work 0.57 0.34 49 RTOG grade 2 AE 2 Gy q.d.
Liver Reference n m TD 50 (Gy) Endpoint Fractionation Scheme Liver Burman et al. 99 0.32 0.5 40 Liver failure.8-2 Gy q.d. Lawrence et al. 992/ 0.69 0.5 45 RTOG grade 3 RILD at 4 months.5-.65 Gy b.i.d. This work Ten Haken et al. 2000 0.994 0.65 42.9 RTOG grade 3 RILD at 4 months.5-.65 Gy b.i.d.; normalized to.5 Gy/fr using α/β of 2 Gy Dawson et al. 200 0.94 0.6 42.0 RTOG grade 3 RILD at 4 months.5-.65 Gy b.i.d.; normalized to.5 Gy/fr using α/β of 2 Gy Dawson et al. 2002 RTOG grade 3 RILD at 4 months.5-.65 Gy b.i.d.; normalized 0.97 0.2 45.8 patients with liver metastases to.5 Gy/fr using α/β of 2 Gy 0.97 0.2 39.8 primary hepatobiliary cancer patients Cheng et al. 2002 0.4 0.26 43 NCI CTC grade 3 RILD at 4 months.8-2 Gy q.d. Cheng et al. 2004 0.35 0.39 49.4 NCI CTC grade 3 RILD at 4 months.8-3 Gy q.d.; normalized to 2 Gy/fr using α/β of 2 Gy Xu et al. 2006 NCI CTC grade 3 RILD at 4 months 4-6 Gy q.d.. 0.7 0.28 0.43 40.5 23 Child-Pugh grade A liver cirrhosis Child-Pugh grade B liver cirrhosis
New LKB Model Parameters Organ n m TD 50 (Gy) Endpoint Fractionation Scheme Lung 0.39 28.6 RTOG grade 2 RP 2 Gy q.d. (0.34, 0.45) (26.2, 3.4) Parotid gland 0.54 29.9 SSF 25% at 3 months 2 Gy q.d. (0.42, 0.75) (26.4, 34.2) Esophagus 0.57 0.34 49 RTOG grade 2 AE 2 Gy q.d. Liver 0.69 0.5 45 RTOG grade 3 RILD at 4 months.5-.65 Gy b.i.d. Optic nerve 0.25 0.4 72 Visual field defects and/or unilateral.8 Gy q.d. blindness Heart 0.636 0.3 50.6 Pericarditis of any grade 2 Gy q.d. Brain stem 0.04 0.5 5.3 Any new or aggravated cranial nerve deficit 0-20 Gy single fraction (radiosurgery) Rectum 0.2 0.5 80 Severe proctitis/necrosis/stenosis/fistula.8-2 Gy q.d. Bladder 0.3 0. 62 RTOG grade 3 bladder complications 2 Gy q.d.
Conclusions Lyman model parameter estimates for nine critical organs identified These estimates are based on post- Emami experience and represent the next iteration in identifying reliable normal tissue complication data for biologically-based based treatment planning