Predictive Model of Liver Toxicity to Aid the Personalized Selection of Proton Versus Photon Therapy in Hepatocellular Carcinoma.

PURPOSE: Our objective was to develop an externally validated model for predicting liver toxicity after radiation therapy in patients with hepatocellular carcinoma (HCC) that can integrate both photon and proton dose distributions with patient-specific characteristics. METHODS AND MATERIALS: Training data consisted of all patients with HCC treated between 2008 and 2019 at our institution (n = 117, 60%/40% photon/proton). We developed a shallow convolutional neural network (CNN) to predict posttreatment liver dysfunction from the differential dose-volume histogram (DVH) and baseline liver metrics. To reduce bias and improve robustness, we used ensemble learning (CNNE). After a preregistered study analysis plan, we evaluated stability using internal bootstrap resampling and generalizability using a data set from a different institution (n = 88). Finally, we implemented a class activation map method to characterize the critical DVH subregions and benchmarked the model against logistic regression and XGBoost. The models were evaluated using the area under the receiver operating characteristic curve and area under the precision-recall curve. RESULTS: The CNNE model showed similar internal performance and robustness compared with the benchmarks. CNNE exceeded the benchmark models in external validation, with an area under the receiver operating characteristic curve of 0.78 versus 0.55 to 0.70, and an area under the precision-recall curve of 0.6 versus 0.43 to 0.52. The model showed improved predictive power in the photon group, excellent specificity in both modalities, and high sensitivity in the photon high-risk group. Models built solely on DVHs confirm outperformance of the CNNE and indicate that the proposed structure efficiently abstracts features from both proton and photon dose distributions. The activation map method demonstrates the importance of the low-dose bath and its interaction with low liver function at baseline. CONCLUSIONS: We developed and externally validated a patient-specific prediction model for hepatic toxicity based on the entire DVH and clinical factors that can integrate both photon and proton therapy cohorts. This model complements the new American Society for Radiation Oncology clinical practice guidelines and could support value-driven integration of proton therapy into the management of HCC.

Toward MR-integrated proton therapy: modeling the potential benefits for liver tumors.

Magnetic resonance imaging (MRI)-integrated proton therapy (MRiPT) is envisioned to improve treatment quality for many cancer patients. However, given the availability of alternative image-guided strategies, its clinical need is yet to be justified. This study aims to compare the expected clinical outcomes of MRiPT with standard of practice cone-beam CT (CBCT)-guided PT, and other MR-guided methods, i.e. offline MR-guided PT and MR-linac, for treatment of liver tumors. Clinical outcomes were assessed by quantifying the dosimetric and biological impact of target margin reduction enabled by each image-guided approach. Planning target volume (PTV) margins were calculated using random and systematic setup, delineation and motion uncertainties, which were quantified by analyzing longitudinal MRI data for 10 patients with liver tumors. Proton treatment plans were created using appropriate PTV margins for each image-guided PT method. Photon plans with margins equivalent to MRiPT were generated to represent MR-linac. Normal tissue complication probabilities (NTCP) of the uninvolved liver were compared. We found that PTV margin can be reduced by 20% and 40% for offline MR-guided PT and MRiPT, respectively, compared with CBCT-guided PT. Furthermore, clinical target volume expansion could be largely alleviated when delineating on MRI rather than CT. Dosimetric implications included decreased equivalent mean dose of the uninvolved liver, i.e. up to 24.4 Gy and 27.3 Gy for offline MR-guided PT and MRiPT compared to CBCT-guided PT, respectively. Considering Child-Pugh score increase as endpoint, NTCP of the uninvolved liver was significantly decreased for MRiPT compared to CBCT-guided PT (up to 48.4%,p < 0.01), offline MR-guided PT (up to 12.9%,p < 0.01) and MR-linac (up to 30.8%,p < 0.05). Target underdose was possible in the absence of MRI-guidance (D90 reduction up to 4.2 Gy in 20% of cases). In conclusion, MRiPT has the potential to significantly reduce healthy liver toxicities in patients with liver tumors. It is superior to other image-guided techniques currently available.

Dosimetric Analysis and Normal-Tissue Complication Probability Modeling of Child-Pugh Score and Albumin-Bilirubin Grade Increase After Hepatic Irradiation.

PURPOSE: This study aimed to develop robust normal-tissue complication probability (NTCP) models for patients with hepatocellular carcinoma treated with radiation therapy (RT) using Child-Pugh (CP) score and albumin-bilirubin (ALBI) grade increase as endpoints for hepatic toxicity. METHODS AND MATERIALS: Data from 108 patients with hepatocellular carcinoma treated with RT between 2008 and 2017 were evaluated, of which 47 patients (44%) were treated with proton RT. Of these patients, 29 received stereotactic body RT and 79 moderately hypofractionated RT to median physical tumor doses of 43 Gy in 5 fractions and 59 Gy in 15 fractions, respectively. A generalized Lyman-Kutcher-Berman (LKB) model was used to model the NTCP using 2 clinical endpoints, both evaluated at 3 months after RT: CP score increase of ≥2 and ALBI grade increase of ≥1 from the pre-RT baseline. Confidence intervals on LKB fit parameters were determined using bootstrap resampling. RESULTS: Compared with previous NTCP models, this study found a stronger correlation between normal liver volume receiving low doses of radiation (5-10 Gy) and a CP score or ALBI grade increase. A CP score increase exhibited a stronger correlation to normal liver volumes irradiated than an ALBI grade increase. LKB models for CP increase found values for the volume-effect parameter of a = 0.06 for all patients, and a = 0.02/0.09 when fit to photon/proton patients separately. Subset analyses for patients with superior initial liver functions showed consistent dose-volume effects (a = 0.1) and consistent dose-response relationships. CONCLUSIONS: This study presents an update of liver NTCP models in the era of modern RT techniques using relevant endpoints of hepatic toxicity, CP score and ALBI grade increase. The results show a stronger influence of low-dose bath on hepatic toxicity than those found in previous studies, indicating that RT techniques that minimize the low-dose bath may be beneficial for patients.

Elective nodal irradiation with simultaneous integrated boost stereotactic body radiotherapy for pancreatic cancer: Analyses of planning feasibility and geometrically driven DVH prediction model.

PURPOSE: We evaluate the feasibility of the elective nodal irradiation strategy in stereotactic body radiotherapy (SBRT) for pancreatic cancer. METHODS: Three simultaneous integrated boost (SIB)-SBRT plans (Boost1, Boost2, and Boost3) were retrospectively generated for each of 20 different patients. Boost1 delivered 33 and 25 Gy to PTV1 and PTV2, respectively. Boost2 delivered 40, 33, and 25 Gy to boostCTV, PTV1, and PTV2, respectively. Boost3 delivered 33 and 25 Gy to PTV1 and PTV3, respectively. PTV1 covered the initial standard SBRT plan (InitPlan) gross tumor volume (GTV). PTV2 covered CTVgeom which was created by a 10-mm expansion (15 mm posterior) of GTV. PTV3 covered CTVprop which included elective nodal regions. The boostCTV included GTV as well as involved vasculature. The planning feasibility in each scenario and dose-volume histograms (DVHs) were analyzed and compared with the InitPlan (delivered 33 Gy only to PTV1) by paired t-test. Next, a novel DVH prediction model was developed and its performance was evaluated according to the prediction accuracy (AC) of planning violations. Then, the model was used to simulate the impacts of GTV-to-organs at risk (OAR) distance and gastrointestinal (GI) OAR volume variations on planning feasibility. RESULTS: Significant dose increases were observed in GI-OARs in SIB-SBRT plans when compared with InitPlan. All dose constraints were met in 63% of cases in InitPlan, Boost1, and Boost2, whereas Boost3 developed DVH violations in all cases. Utilizing previous patient anatomy, the novel DVH prediction model achieved a high AC in the prediction of violations for GI-OARs; the positive predictive value, negative predictive value, and AC were 66%, 90%, and 84%, respectively. Experiments with the model demonstrated that the larger proximity volume of GI-OAR at the shorter distance substantially impacted on planning violations. CONCLUSIONS: SIB-SBRT plan with geometrically defined prophylactic areas can be dosimetrically feasible, but including all nodal areas with 25 Gy in five fractions appears to be unrealistic.

Differential Association Between Circulating Lymphocyte Populations With Outcome After Radiation Therapy in Subtypes of Liver Cancer.

PURPOSE: Irradiation may have significant immunomodulatory effects that impact tumor response and could potentiate immunotherapeutic approaches. The purposes of this study were to prospectively investigate circulating lymphoid cell population fractions during hypofractionated proton therapy (HPT) in blood samples of liver cancer patients and to explore their association with survival. METHODS AND MATERIALS: We collected serial blood samples before treatment and at days 8 and 15 of HPT from 43 patients with liver cancer-22 with hepatocellular carcinoma (HCC) and 21 with intrahepatic cholangiocarcinoma (ICC)-enrolled in a phase 2 clinical trial. All patients received 15 fractions of proton therapy to a median dose of 58 Gy (relative biological effectiveness). We used flow cytometry to measure the changes in the fractions of total CD3+, CD4+, and CD8+ T cells; CD4+ CD25+ T cells; CD4+ CD127+ T cells; CD3+ CD8+ CD25+ activated cytotoxic T lymphocytes (CTLs); and CD3- CD56+ natural killer cells. RESULTS: With a median follow-up period of 42 months, median overall survival (OS) in the study cohort was 30.6 months for HCC and 14.5 months for ICC patients. Longer OS was significantly correlated with greater CD4+ CD25+ T-cell (P = .003) and CD4+ CD127+ T-cell (P = .01) fractions at baseline only in ICC patients. In HCC patients, the fraction of activated CTLs mid treatment (at day 8) was significantly associated with OS (P = .007). These findings suggest a differential relevance of immunomodulation by HPT in these liver cancers. CONCLUSIONS: Antitumor immunity may depend on maintenance of a sufficiently high number of activated CTLs during HPT in HCC patients and CD4+ CD25+ T cells and CD4+ CD127+ T cells prior to treatment in ICC patients. These results could guide the design of future studies to determine the optimal treatment schedules when combining irradiation with specific immunotherapy approaches.

Liver reirradiation for patients with hepatocellular carcinoma and liver metastasis.

PURPOSE: This study aimed to assess the safety and efficacy of administering liver reirradiation to patients with primary liver tumors or liver metastasis. METHODS AND MATERIALS: A total of 49 patients (with 64 individual tumors) who received liver reirradiation at our institution between June 2008 and December 2016 were identified for retrospective review. Patients were treated to the same, different, or a combination of previously treated liver tumors for recurrent primary (53%) or metastatic (47%) disease using photons or protons. Clinical and treatment-related factors were compiled and patients were monitored for toxicity and evidence of classic or nonclassic radiation-induced liver disease. Survival was estimated with the Kaplan-Meier method and cumulative incidence of local failure (LF) was used to estimate LF using the Response Evaluation Criteria in Solid Tumors version 1.1. RESULTS: The median age at the time of reirradiation was 72 years and the median interval between radiation courses was 9 months. At a median follow-up of 10.5 months, 36 patients (73%) had died, 9 patients (18%) were alive, and 4 patients (8%) were lost to follow-up. The median survival for the cohort was 14 months. The overall 1-year estimate of LF was 46.4%. The 1-year estimates of LF for liver metastases and hepatocellular carcinoma were 61.0% and 32.5%, respectively. The average prescription dose was similar between the reirradiation and initial courses (equivalent dose in 2 Gy fractions EQD2: 65.0 vs 64.3 Gyα/ß = 10, respectively) but the average dose to the untreated liver was lower at the time of reirradiation (EQD2: 10.5 vs 13.9 Gyα/ß = 3, respectively, P = .01). Among patients with hepatocellular carcinoma, the average normal liver dose was significantly larger for patients who exhibited a worsening of Child-Pugh score after reirradiation compared with those who did not (1210 cGy vs 759 cGy, P = .04). With regard to toxicity, 85.7% of patients experienced grade 1 to 2 toxicity, 4.1% developed grade 3, and only 2 patients (4.1%) met the criteria for radiation-induced liver disease after reirradiation. CONCLUSIONS: Liver reirradiation may be an effective and safe option for select patients; however, further prospective study is necessary to establish treatment guidelines and recommended dosing.

Phase II Study of Proton-Based Stereotactic Body Radiation Therapy for Liver Metastases: Importance of Tumor Genotype.

Background: We evaluated the efficacy and safety of risk-adapted, proton-based stereotactic body radiation therapy (SBRT) for liver metastases from solid tumors. Methods: This single-arm phase II single institutional study (NCT01239381) included patients with limited extrahepatic disease, 800 mL or greater of uninvolved liver, and no cirrhosis or Child-Pugh A, who had received proton-based SBRT to one to four liver metastases from solid tumors. Treatment comprised 30 to 50 Gray equivalent (GyE) in five fractions based on the effective volume of liver irradiated. Sample size was calculated to determine if local control (LC) at one year was greater than 70%. The cumulative incidence of local failure was used to estimate LC. The association of tumor characteristics, including genetic alterations in common cancer genes such as BRAF, EGFR, HER2, KRAS, NRAS, PIK3CA, and TP53 with local tumor control, was assessed. All statistical tests were two-sided. Results: Eighty-nine patients were evaluable (colorectal, n = 34; pancreatic, n = 13; esophagogastric, n = 12; other, n = 30). Median tumor size was 2.5 cm (range = 0.5-11.9 cm). Median dose was 40 GyE (range = 30-50 GyE), and median follow-up was 30.1 months (range = 14.7-53.8 months). There was no grade 3 to 5 toxicity. Median survival time was 18.1 months. The one- and three-year LC rates were 71.9% (95% confidence limit [CL] = 62.3% to 80.9%) and 61.2% (95% CL = 50.8% to 71.8%), respectively. For large tumors (≥6 cm), one-year LC remained high at 73.9% (95% CL = 54.6% to 89.8%). Mutation in the KRAS oncogene was the strongest predictor of poor LC (P = .02). Tumor with both mutant KRAS and TP53 were particularly radioresistant, with a one-year LC rate of only 20.0%, compared with 69.2% for all others (P = .001). Conclusions: We report the largest prospective evaluation to date of liver SBRT for hepatic metastases, and the first with protons. Protons were remarkably well tolerated and effective even for metastases that were 6 cm or larger. KRAS mutation is a strong predictor of poor LC, stressing the need for tumor genotyping prior to SBRT and treatment intensification in this patient subset.

Gallbladder toxicity and high-dose ablative-intent radiation for liver tumors: Should we constrain the dose?

PURPOSE: Little is known about the risk of gallbladder toxicity from hypofractionated (HFXRT) and stereotactic body radiation therapy (SBRT). We report on gallbladder toxicity and attribution to treatment in a prospective series of patients with primary and metastatic liver tumors receiving ablative-intent HFXRT and SBRT with protons. METHODS AND MATERIALS: We evaluated 93 patients with intact gallbladders enrolled in either of 2 trials investigating proton HFXRT and SBRT for primary and metastatic liver tumors from 2009 to 2014. Patients received 45 to 67.5 GyE in 15 fractions for primary liver tumors (n = 45) and 30 to 50 GyE in 5 fractions for metastatic tumors (n = 48). No gallbladder dose constraints were used at treatment, and gallbladder volumes and dose-volume histograms were created retrospectively. Attributable toxicity was defined as cholecystitis or perforation without preexisting gallbladder disease. Baseline factors were evaluated using Fisher exact test and the nonparametric K-sample test. RESULTS: At baseline, 25 patients had preexisting cholelithiasis and 15 underwent biliary stenting before or after RT. Median follow-up after treatment was 11.8 months (range, 0.1-59.2 months). Despite maximum gallbladder doses >70 GyE in 41%, >80 GyE in 31%, and >90 GyE in 13% (equieffective dose at 2 Gy [EQD2], α/ß = 3), there were no attributable cases of gallbladder toxicity. Two patients developed grade 3 and 4 cholecystitis 16 and 2 months after treatment, respectively, and both had a strong history of preexisting cholelithiasis and biliary stenting. These patients received relatively low gallbladder doses with mean doses of 0.02 GyE and 5.1 GyE (EQD2, α/ß = 3), well below the 17.1 GyE mean for the remaining cohort (range, 0-81.1 GyE, EQD2). CONCLUSIONS: We identified no relationship between gallbladder dose and toxicity and did not reach the maximum tolerated gallbladder dose in this cohort treated with high-dose radiation. We recommend not constraining dose to the gross tumor volume to protect the gallbladder during ablative HFXRT and SBRT.

Impact of intravenous contrast enhancement phase on target definition for hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (IHC): Observations from patients enrolled on a prospective phase 2 trial.

BACKGROUND: The safety and efficacy of radiation therapy for hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (IHC) depend on accurate definition of gross tumor volume (GTV), but GTV often varies between phases of multiphasic computed tomography (CT) imaging. METHODS: We contoured GTVs on arterial, portal venous, and delayed phases of multiphasic CT scans for 32 patients treated on an institutional review board-approved prospective trial of proton therapy for primary liver tumors and determined which phase provided optimal GTV visualization. We assessed agreement between individual phase GTVs to determine if GTV for each phase was encompassed in a 5-mm expansion of either the smallest or the best-visualized GTV. RESULTS: There were 19 HCC lesions and 14 IHC lesions. HCC lesions were best identified on the arterial phase in 42% (n = 8), portal venous phase in 32% (n = 6), and delayed phase in 26% (n = 5). IHC lesions were best identified on portal venous phase in 64% (n = 9) and the arterial phase in 29% (n = 4), with 1 case equally visualized on arterial and portal venous phases. In all 33 lesions, a 5-mm expansion around the smallest GTV failed to cover GTVs defined on other available phases. A 5-mm expansion around the best-visualized GTV provided satisfactory coverage of all available phases' GTVs in 6/18 HCC cases and 2/9 IHC cases. CONCLUSIONS: Variability between GTVs on multiphasic CT scans could not be overcome with a 5-mm expansion of either the smallest GTV or the best-visualized GTV. Assessment of all available intravenous contrast phases is essential to accurately define the GTV.

Multi-Institutional Phase II Study of High-Dose Hypofractionated Proton Beam Therapy in Patients With Localized, Unresectable Hepatocellular Carcinoma and Intrahepatic Cholangiocarcinoma.

PURPOSE: To evaluate the efficacy and safety of high-dose, hypofractionated proton beam therapy for hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC). MATERIALS AND METHODS: In this single-arm, phase II, multi-institutional study, 92 patients with biopsy-confirmed HCC or ICC, determined to be unresectable by multidisciplinary review, with a Child-Turcotte-Pugh score (CTP) of A or B, ECOG performance status of 0 to 2, no extrahepatic disease, and no prior radiation received 15 fractions of proton therapy to a maximum total dose of 67.5 Gy equivalent. Sample size was calculated to demonstrate > 80% local control (LC) defined by Response Evaluation Criteria in Solid Tumors (RECIST) 1.0 criteria at 2 years for HCC patients, with the parallel goal of obtaining acceptable precision for estimating outcomes for ICC. RESULTS: Eighty-three patients were evaluable: 44 with HCC, 37 with ICC, and two with mixed HCC/ICC. The CTP score was A for 79.5% of patients and B for 15.7%; 4.8% of patients had no cirrhosis. Prior treatment had been given to 31.8% of HCC patients and 61.5% of ICC patients. The median maximum dimension was 5.0 cm (range, 1.9 to 12.0 cm) for HCC patients and 6.0 cm (range, 2.2 to 10.9 cm) for ICC patients. Multiple tumors were present in 27.3% of HCC patients and in 12.8% of ICC patients. Tumor vascular thrombosis was present in 29.5% of HCC patients and in 28.2% of ICC patients. The median dose delivered to both HCC and ICC patients was 58.0 Gy. With a median follow-up among survivors of 19.5 months, the LC rate at 2 years was 94.8% for HCC and 94.1% for ICC. The overall survival rate at 2 years was 63.2% for HCC and 46.5% ICC. CONCLUSION: High-dose hypofractionated proton therapy demonstrated high LC rates for HCC and ICC safely, supporting ongoing phase III trials of radiation in HCC and ICC.

Deep inspiration breath-hold technique in left-sided breast cancer radiation therapy: Evaluating cardiac contact distance as a predictor of cardiac exposure for patient selection.

PURPOSE: The purpose of this study was to evaluate the efficacy of voluntary deep inspiration breath-hold (DIBH) over a free-breathing (FB) technique to minimize cardiac radiation exposure in radiation therapy of left-sided breast cancer. Also, to better select patients for DIBH, the correlation between cardiac contact distance (CCD) and cardiac dose was assessed. METHODS AND MATERIALS: Thirty-five patients with left-sided breast cancer underwent DIBH and FB planning computed tomography scans, and the 2 plans were compared. Dose-volume histograms were analyzed for heart, left anterior descending coronary artery (LAD), left ventricle (LV), and left lung. Axial CCDs and parasagittal CCDs (FB-CCDps) were measured on FB planning computed tomography scans. RESULTS: Dose to heart, LAD, LV, and left lung was significantly lower in DIBH plans than in FB by all metrics. When DIBH was compared with FB, mean dose (Dmean) for heart was 0.9 versus 2.5 Gy; for LAD, 4.0 versus 14.9 Gy; and for LV, 1.1 versus 3.9 Gy (P < .0001), respectively. Seventy-five percent of the patients had a dose reduction of ≥ 0.9 Gy in Dmean to heart, ≥ 3 Gy in Dmean to LAD, and ≥ 1.7 Gy in Dmean to LV. FB-CCDps was associated with an equivalent uniform dose to heart, LAD, and LV for both the DIBH and FB plans (P ≤ .01); FB axial CCD measures were not. CONCLUSIONS: DIBH is a simple and highly effective technique to reduce cardiac exposure without compromising target coverage. FB-CCDps is potentially a very good predictor for cardiac exposure: the longer the FB-CCDps, the higher the dose. Our findings suggest that at least 75% of patients with left-sided breast cancer might benefit from the DIBH technique in terms of potentially clinically relevant dose reduction to cardiac structures, and therefore, it should be instituted as routine clinical practice.

A prospective feasibility study of respiratory-gated proton beam therapy for liver tumors.

PURPOSE: To evaluate the feasibility of a respiratory-gated proton beam therapy for liver tumors. METHODS AND MATERIALS: Fifteen patients were enrolled in a prospective institutional review board-approved protocol. Eligibility criteria included Childs-Pugh A/B cirrhosis, unresectable biopsy- proven hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC), or metastatic disease (solid tumors only), 1-3 lesions, and tumor size of ≤6 cm. Patients received 15 fractions to a total dose of 45-75 GyE [gray equivalent] using respiratory-gated proton beam therapy. Gating was performed with an external respiratory position monitoring based system. RESULTS: Of the 15 patients enrolled in this clinical trial, 11 had HCC, 3 had ICC, and 1 had metastasis from another primary. Ten patients had a single lesion, 3 patients had 2 lesions, and 2 patients had 3 lesions. Toxicities were grade 3 bilirubinemia-2, grade 3 gastrointestinal bleed-1, and grade 5 stomach perforation-1. One patient had a marginal recurrence, 3 had hepatic recurrences elsewhere in the liver, and 2 had extrahepatic recurrence. With a median follow-up for survivors of 69 months, 1-, 2-, and 3-year overall survivals are 53%, 40%, and 33%, respectively. Progression-free survivals are 40%, 33%, and 27% at 1, 2, and 3 years, respectively. CONCLUSIONS: Respiratory-gated proton beam therapy for liver tumors is feasible. Phase 2 studies for primary liver tumors and metastatic tumors are underway.

Variations in tumor size and position due to irregular breathing in 4D-CT: a simulation study.

PURPOSE: To estimate the position and volume errors in 4D-CT caused by irregular breathing. METHODS: A virtual 4D-CT scanner was designed to reproduce axial mode scans with retrospective resorting. This virtual scanner creates an artificial spherical tumor based on the specifications of the user, and recreates images that might be produced by a 4D-CT scanner using a patient breathing waveform. 155 respiratory waveforms of patients were used to test the variability of 4D-CT scans. Each breathing waveform was normalized and scaled to 1, 2, and 3 cm peak-to-peak motion, and artificial tumors with 2 and 4 cm radius were simulated for each scaled waveform. The center of mass and volume of resorted 4D-CT images were calculated and compared to the expected values of center of mass and volume for the artificial tumor. Intrasubject variability was investigated by running the virtual scanner over different subintervals of each patient's breathing waveform. RESULTS: The average error in the center of mass location of an artificial tumor was less than 2 mm standard deviation for 2 cm motion. The corresponding average error in volume was less than 4%. In the worst-case scenarios, a center of mass error of 1.0 cm standard deviation and volume errors of 30%-60% at inhale were found. Systematic errors were observed in a subset of patients due to irregular breathing, and these errors were more pronounced when the tumor volume is smaller. CONCLUSIONS: Irregular breathing during 4D-CT simulation causes systematic errors in volume and center of mass measurements. These errors are small but depend on the tumor size, motion amplitude, and degree of breathing irregularity.

An effective preoperative three-dimensional radiotherapy target volume for extremity soft tissue sarcoma and the effect of margin width on local control.

PURPOSE: There is little information on the appropriate three-dimensional (3D) preoperative radiotherapy (XRT) volume for extremity soft-tissue sarcomas (STS). We retrospectively analyzed the pattern of local failure (LF) to help elucidate optimal field design. METHODS AND MATERIALS: We analyzed the 56 patients who underwent computed tomography-planned XRT for Stage I to III extremity STS between June 2000 and December 2006. Clinical target volume (CTV) included the T1 post-gadolinium-defined gross tumor volume with 1- to 1.5-cm radial and 3.5-cm longitudinal margins. Planning target volume expansion was 5 to 7 mm, and >or=95% of dose was delivered to the planning target volume. Preoperative XRT was 44 to 50.4 Gy (median, 50). Postoperative boost of 10 to 20 Gy was given to 12 patients (6 with positive and 6 with close margins). RESULTS: Follow-up ranged from 15 to 76 months (median, 41 months). The 5-year local control, freedom from distant metastasis, disease-free survival, and overall survival were 88.5%, 80.0%, 77.5% and 82.8%, respectively. Three patients (all with positive margin) experienced local failure (LF) as first relapse (2 isolated, 1 with distant failure), and 2 additional patients (all with margin<1 mm) had late LF after distant metastasis. The LFs were within the CTV in 3 patients and within and also extending beyond the CTV in 2 patients. CONCLUSIONS: These target volume definitions appear to be appropriate for most patients. No local recurrences were observed with surgical margins >or=1 mm, and it appears that these may be adequate for patients with extremity STS treated with preoperative radiotherapy.

Effects of interfractional anatomical changes on water-equivalent pathlength in charged-particle radiotherapy of lung cancer.

Intrafractional motion and interfractional changes affect the accuracy of the delivered dose in radiotherapy, particularly in charged-particle radiotherapy. Most recent studies are focused on intrafractional motion (respiratory motion). Here, we report a quantitative simulation analysis of the effects of interfractional changes on water-equivalent pathlength (WEL) in charged-particle lung therapy. Serial four-dimensional (4D) CT scans were performed under free breathing conditions; the time span between the first and second 4DCT scans was five weeks. We quantified WEL changes between the first and second CT scans due to interfractional changes (tumor shrinkage and tissue density changes) and compared the particle-beam-stopping point between the serial 4DCT scans with use of the same initial bolus. Both tumor-shrinkage and lung-density changes were observed in a single patient over the course of therapy. The lung density decreased by approximately 0.1 g/cm(3) between the first and second-CT scans, resulting in a 1.5 cm WEL changes. Tumor shrinkage resulted in approximately 3 cm WEL changes. If the same initial bolus and plan were used through the treatment course, an unexpected significant beam overshoot would occur by interfractional changes due to tumor shrinkage and lung density variation.

Implications of respiratory motion as measured by four-dimensional computed tomography for radiation treatment planning of esophageal cancer.

PURPOSE: To evaluate the respiratory motion of primary esophageal cancers and pathologic celiac-region lymph nodes using time-resolved four-dimensional computed tomography (4D CT). METHODS AND MATERIALS: Respiration-synchronized 4D CT scans were obtained to quantify the motion of primary tumors located in the proximal, mid-, or distal thoracic esophagus, as well as any involved celiac-region lymph nodes. Respiratory motion was measured in the superior-inferior (SI), anterior-posterior (AP), and left-right (LR) directions and was analyzed for correlation with anatomic location. Recommended margin expansions were determined for both primary and nodal targets. RESULTS: Thirty patients underwent 4D CT scans at Massachusetts General Hospital for planned curative treatment of esophageal cancer. Measurements of respiratory tumor motion were obtained for 1 proximal, 4 mid-, and 25 distal esophageal tumors, as well as 12 involved celiac-region lymph nodes. The mean (SD) peak-to-peak displacements of all primary tumors in the SI, AP, and LR dimensions were 0.80 (0.45) cm, 0.28 (0.20) cm, and 0.22 (0.23) cm, respectively. Distal tumors were found to have significantly greater SI and AP motion than proximal or mid-esophageal tumors. The mean (SD) SI, AP, and LR peak-to-peak displacements of the celiac-region lymph nodes were 0.92 (0.56) cm, 0.46 (0.27) cm, and 0.19 (0.26) cm, respectively. CONCLUSIONS: Margins of 1.5 cm SI, 0.75 cm AP, and 0.75 cm LR would account for respiratory tumor motion of >95% of esophageal primary tumors in the dataset. All celiac-region lymph nodes would be adequately covered with SI, AP, and LR margins of 2.25 cm, 1.0 cm, and 0.75 cm, respectively.

Quantification of mediastinal and hilar lymph node movement using four-dimensional computed tomography scan: implications for radiation treatment planning.

PURPOSE: To quantitatively describe mediastinal and hilar lymph node movement in patients with lymph node-positive lung cancer. METHODS AND MATERIALS: Twenty-four patients with lung cancer who underwent four-dimensional computed tomography scanning at Massachusetts General Hospital were included in the study. The maximum extent of superior motion of the superior border was measured, as well as the maximum inferior movement of the inferior border. The average of these two values is defined as the peak-to-peak movement. This process was repeated for mediolateral (ML) and anterior-posterior (AP) movement. Linear regression was used to determine lymph node characteristics associated with peak-to-peak movement. Various uniform expansions were investigated to determine the expansion margins necessary to ensure complete internal target volume (ITV) coverage. RESULTS: The mean peak-to-peak displacements of paratracheal lymph nodes were 4 mm (craniocaudal [CC]), 2 mm (ML), and 2 mm (AP). For subcarinal lymph nodes, the mean peak-to-peak movements were 6 mm (CC), 4 mm (ML), and 2 mm (AP). The mean peak-to-peak displacements of hilar lymph nodes were 7 mm (CC), 1 mm (ML), and 4 mm (AP). On multivariate analysis, lymph node station and lymph node size were significantly related to peak-to-peak movement. Expansions of 8 mm for paratracheal nodes and 13 mm for subcarinal and hilar nodes would have been necessary to cover the ITV of 95% of these nodal masses. CONCLUSIONS: Subcarinal and hilar lymph nodes may move substantially throughout the respiratory cycle. In the absence of patient-specific information on nodal motion, expansions of at least 8 mm, 13 mm, and 13 mm should be considered to cover the ITV of paratracheal, subcarinal, and hilar lymph nodes, respectively.

Design of 4D treatment planning target volumes.

PURPOSE: When using non-patient-specific treatment planning margins, respiratory motion may lead to geometric miss of the target while unnecessarily irradiating normal tissue. Imaging different respiratory states of a patient allows patient-specific target design. We used four-dimensional computed tomography (4DCT) to characterize tumor motion and create treatment volumes in 10 patients with lung cancer. These were compared with standard treatment volumes. METHODS AND MATERIALS: Four-dimensional CT and free breathing helical CT data of 10 patients were acquired. Gross target volumes (GTV) were delineated on the helical scan as well as on each phase of the 4D data. Composite GTVs were defined on 4DCT. Planning target volumes (PTV) including clinical target volume, internal margin (IM), and setup margin were generated. 4DPTVs with different IMs and standard PTVs were compared by computing centroid positions, volumes, volumetric overlap, and bounding boxes. RESULTS: Four-dimensional PTVs and conventional PTVs differed in volume and centroid positions. Overlap between 4DPTVs generated from two extreme tumor positions only compared with 10 respiratory phases was 93.7%. Comparing PTVs with margins of 15 mm (IM 5 mm) on composite 4D target volumes to PTVs with 20 mm (IM 10 mm) on helical CT data resulted in a decrease in target volume sizes by 23% on average. CONCLUSION: With patient-specific characterization of tumor motion, it should be possible to decrease internal margins. Patient-specific treatment volumes can be generated using extreme tumor positions on 4DCT. To date, more than 150 patients have been treated using 4D target design.

Mapping of nodal disease in locally advanced prostate cancer: rethinking the clinical target volume for pelvic nodal irradiation based on vascular rather than bony anatomy.

PURPOSE: Toxicity from pelvic irradiation could be reduced if fields were limited to likely areas of nodal involvement rather than using the standard "four-field box." We employed a novel magnetic resonance lymphangiographic technique to highlight the likely sites of occult nodal metastasis from prostate cancer. METHODS AND MATERIALS: Eighteen prostate cancer patients with pathologically confirmed node-positive disease had a total of 69 pathologic nodes identifiable by lymphotropic nanoparticle-enhanced MRI and semiquantitative nodal analysis. Fourteen of these nodes were in the para-aortic region, and 55 were in the pelvis. The position of each of these malignant nodes was mapped to a common template based on its relation to skeletal or vascular anatomy. RESULTS: Relative to skeletal anatomy, nodes covered a diffuse volume from the mid lumbar spine to the superior pubic ramus and along the sacrum and pelvic side walls. In contrast, the nodal metastases mapped much more tightly relative to the large pelvic vessels. A proposed pelvic clinical target volume to encompass the region at greatest risk of containing occult nodal metastases would include a 2.0-cm radial expansion volume around the distal common iliac and proximal external and internal iliac vessels that would encompass 94.5% of the pelvic nodes at risk as defined by our node-positive prostate cancer patient cohort. CONCLUSIONS: Nodal metastases from prostate cancer are largely localized along the major pelvic vasculature. Defining nodal radiation treatment portals based on vascular rather than bony anatomy may allow for a significant decrease in normal pelvic tissue irradiation and its associated toxicities.