Chang-Yu Wang

The treatment responses in cases of radiation therapy to portal vein thrombosis in advanced hepatocellular carcinoma.

PURPOSE To review the response to radiation therapy for hepatocellular carcinoma (HCC) with portal vein thrombosis (PVT) and determine the factors favoring its efficacy. METHODS AND MATERIALS Patients with HCC and PVT referred for radiation therapy between 1997 and 2005 were retrospectively reviewed. Patients who had undergone treatment to primary HCC before radiation or had extrahepatic metastasis were excluded. A radiation dose of 60 Gy with 2 to 3Gy per fraction was prescribed. Clinical features before therapy were investigated, and the most significant imaging change after radiotherapy was regarded as the treatment response. Survival times were compared and the hazard ratios of independent variables were determined. RESULTS The treatment response rate of the 326 patients included in the study was 25.2% (n = 82). The median survival times were 13.3, 11.6, 9.0, 4.5, and 2.1 months for complete response, partial response, vascular transformation, no response, and the lost follow-up patients, respectively. Statistically significant differences in survival were not found among responder groups (p = 0.224-0.916) but were found between responders and nonresponders (p = 0.002). The most significant independent variables associated with survival (p < 0.001) were performance status and radiation dose. Minor independent factors were ascites, alfa-fetoprotein, albumin, and HBsAg (p = 0.009-0.038). In patients with favorable performance status, those with no more than one minor risk factor had a superior prognosis after radiation therapy (p = 0.013). This result was verified by a review of similar patients in 2006. CONCLUSION Radiation therapy is the treatment of choice for selected HCC patients with PVT.

Multivariate Analysis of Para-Aortic Lymph Node Recurrence After Definitive Radiotherapy for Stage IB-IVA Squamous Cell Carcinoma of Uterine Cervix

PURPOSE To evaluate the pretreatment risk factors of para-aortic lymph node (PALN) recurrence after primary radiotherapy for cervical cancer. METHODS AND MATERIALS Between May 1992 and January 2006, the data from 758 patients with squamous cell carcinoma of the uterine cervix were retrospectively analyzed. No patient had undergone PALN radiotherapy as their initial treatment. PALN recurrence was diagnosed by computed tomography. PALN relapse-free status was determined clinically or radiographically. We analyzed the actuarial rates of PALN recurrence using Kaplan-Meier curves. Multivariate analyses were performed with Cox regression models. RESULTS Of the 758 patients, 38 (5%) and 42 (6%) had isolated and nonisolated PALN recurrences after a median follow-up of 50 months (range, 2-159 months), respectively. The 3-year and 5-year overall survival rate after PALN recurrence was 35% and 28%, respectively. A squamous cell carcinoma antigen (SCC-Ag) level >40 ng/mL (p <0.001), advanced parametrial involvement (score 4-6; p = 0.002), and the presence of pelvic lymphadenopathy (p = 0.007) were independent factors associated with PALN relapse on multivariate analysis. The 5-year PALN recurrence rate in patients with a SCC-Ag level >40 ng/mL, SCC-Ag level of 20-40 ng/mL, parametrial score of 4-6, pelvic lymphadenopathy, and no risk factors was 57%, 22%, 34%, 37%, and 9%, respectively. CONCLUSIONS Patients with squamous cell carcinoma of the uterine cervix and a high SCC-Ag level, pelvic lymphadenopathy, or advanced PM involvement were predisposed to PALN recurrence after definitive radiotherapy. More intensive follow-up schedules are suggested for early detection and salvage in high-risk patients.

Dosimetric Comparison of Helical Tomotherapy and Dynamic Conformal Arc Therapy in Stereotactic Radiosurgery for Vestibular Schwannomas

The dosimetric results of stereotactic radiosurgery (SRS) for vestibular schwannoma (VS) performed using dynamic conformal arc therapy (DCAT) with the Novalis system and helical TomoTherapy (HT) were compared using plan quality indices. The HT plans were created for 10 consecutive patients with VS previously treated with SRS using the Novalis system. The dosimetric indices used to compare the techniques included the conformity index (CI) and homogeneity index (HI) for the planned target volume (PTV), the comprehensive quality index (CQI) for nine organs at risk (OARs), gradient score index (GSI) for the dose drop-off outside the PTV, and plan quality index (PQI), which was verified using the plan quality discerning power (PQDP) to incorporate 3 plan indices, to evaluate the rival plans. The PTV ranged from 0.27-19.99 cm(3) (median 3.39 cm(3)), with minimum required PTV prescribed doses of 10-16 Gy (median 12 Gy). Both systems satisfied the minimum required PTV prescription doses. HT conformed better to the PTV (CI: 1.51 ± 0.23 vs. 1.94 ± 0.34; p < 0.01), but had a worse drop-off outside the PTV (GSI: 40.3 ± 10.9 vs. 64.9 ± 13.6; p < 0.01) compared with DCAT. No significant difference in PTV homogeneity was observed (HI: 1.08 ± 0.03 vs. 1.09 ± 0.02; p = 0.20). HT had a significantly lower maximum dose in 4 OARs and significant lower mean dose in 1 OAR; by contrast, DCAT had a significantly lower maximum dose in 1 OAR and significant lower mean dose in 2 OARs, with the CQI of the 9 OARs = 0.92 ± 0.45. Plan analysis using PQI (HT 0.37 ± 0.12 vs. DCAT 0.65 ± 0.08; p < 0.01), and verified using the PQDP, confirmed the dosimetric advantage of HT. However, the HT system had a longer beam-on time (33.2 ± 7.4 vs. 4.6 ± 0.9 min; p < 0.01) and consumed more monitor units (16772 ± 3803 vs. 1776 ± 356.3; p < 0.01). HT had a better dose conformity and similar dose homogeneity but worse dose gradient than DCAT. Plan analysis confirmed the dosimetric advantage of HT, although not all indices revealed a better outcome for HT. Whether this dosimetric advantage translates into a clinical benefit deserves further investigation.

Fuzzy Logic-Based Prognostic Score for Outcome Prediction in Esophageal Cancer

Given the poor prognosis of esophageal cancer and the invasiveness of combined modality treatment, improved prognostic scoring systems are needed. We developed a fuzzy logic-based system to improve the predictive performance of a risk score based on the serum concentrations of C-reactive protein (CRP) and albumin in a cohort of 271 patients with esophageal cancer before radiotherapy. Univariate and multivariate survival analyses were employed to validate the independent prognostic value of the fuzzy risk score. To further compare the predictive performance of the fuzzy risk score with other prognostic scoring systems, time-dependent receiver operating characteristic curve analysis was used. Application of fuzzy logic to the serum values of CRP and albumin increased predictive performance for one-year overall survival (AUC = 0.773) compared with that of a single marker (AUC = 0.743 and 0.700 for CRP and albumin, respectively), where the AUC denotes the area under curve. This fuzzy logic-based approach also performed consistently better than the Glasgow prognostic score (AUC = 0.745). Thus, application of fuzzy logic to the analysis of serum markers can more accurately predict the outcome for patients with esophageal cancer.

Pretreatment carcinoembryonic antigen level is a risk factor for para-aortic lymph node recurrence in addition to squamous cell carcinoma antigen following definitive concurrent chemoradiotherapy for squamous cell carcinoma of the uterine cervix

BackgroundTo identify pretreatment carcinoembryonic antigen (CEA) levels as a risk factor for para-aortic lymph node (PALN) recurrence following concurrent chemoradiotherapy (CCRT) for cervical cancer.MethodsFrom March 1995 to January 2008, 188 patients with squamous cell carcinoma (SCC) of the uterine cervix were analyzed retrospectively. No patient received PALN irradiation as the initial treatment. CEA and squamous cell carcinoma antigen (SCC-Ag) were measured before and after radiotherapy. PALN recurrence was detected by computer tomography (CT) scans. We analyzed the actuarial rates of PALN recurrence by using Kaplan-Meier curves. Multivariate analyses were carried out with Cox regression models. We stratified the risk groups based on the hazard ratios (HR).ResultsBoth pretreatment CEA levels ≥ 10 ng/mL and SCC-Ag levels < 10 ng/mL (p < 0.001, HR = 8.838), SCC-Ag levels ≥ 40 ng/mL (p < 0.001, HR = 12.551), and SCC-Ag levels of 10-40 ng/mL (p < 0.001, HR = 4.2464) were significant factors for PALN recurrence. The corresponding 5-year PALN recurrence rates were 51.5%, 84.8%, and 27.5%, respectively. The 5-year PALN recurrence rate for patients with both low (< 10 ng/mL) SCC and CEA was only 9.6%. CEA levels ≥ 10 ng/mL or SCC-Ag levels ≥ 10 ng/mL at PALN recurrence were associated with overall survival after an isolated PALN recurrence. Pretreatment CEA levels ≥ 10 ng/mL were also associated with survival after an isolated PALN recurrence.ConclusionsPretreatment CEA ≥ 10 ng/mL is an additional risk factor of PALN relapse following definitive CCRT for SCC of the uterine cervix in patients with pretreatment SCC-Ag levels < 10 ng/mL. More comprehensive examinations before CCRT and intensive follow-up schedules are suggested for early detection and salvage in patients with SCC-Ag or CEA levels ≥ 10 ng/mL.

Prognostic Value of Pretreatment Carcinoembryonic Antigen After Definitive Radiotherapy With or Without Concurrent Chemotherapy for Squamous Cell Carcinoma of the Uterine Cervix

PURPOSE To evaluate whether pretreatment carcinoembryonic antigen (CEA) levels have a prognostic role in patients after definitive radiotherapy for squamous cell carcinoma (SCC) of the uterine cervix. METHODS AND MATERIALS A retrospective study of 550 patients was performed. The SCC antigen (SCC-Ag) and CEA levels were regarded as elevated when they were ≥2 and ≥5 ng/mL, respectively. A total of 208 patients underwent concurrent chemoradiotherapy (CCRT). The Kaplan-Meier method was used to calculate the distant metastasis (DM), local failure (LF), disease-free survival (DFS), and overall survival (OS) rates. Multivariate analysis was performed using the Cox proportional hazards model. The hazard ratio (HR) with 95% confidence interval (CI) was evaluated for the risk of a poor prognosis. RESULTS Compared with the patients with normal CEA/SCC-Ag levels, CEA levels ≥10 ng/mL but without elevated SCC-Ag levels was an independent factor for LF (HR, 51.81; 95% CI, 11.51-233.23; p < .001), DM (HR, 6.04; 95% CI, 1.58-23.01; p = .008), DFS (HR, 10.17; 95% CI, 3.18-32.56; p < .001), and OS (HR, 5.75; 95% CI, 1.82-18.18; p = .003) after RT alone. However, no significant role for CEA was noted in patients with SCC-Ag levels ≥2 ng/mL. In patients undergoing CCRT, a CEA level ≥10 ng/mL was an independent factor for LF (HR, 2.50; 95% CI, 1.01-6.21; p = .047), DM (HR, 3.41; 95% CI, 1.56-7.46; p = .002), DFS (HR, 2.73; 95% CI, 1.39-5.36; p = .003), and OS (HR, 3.93; 95% CI 1.99-7.75; p < .001). A SCC-Ag level of ≥40 ng/mL was another prognostic factor for DM, DFS, and OS in patients undergoing not only CCRT, but also RT alone. The 5-year OS rate for CCRT patients with CEA <10 ng/mL and ≥10 ng/mL was 75.3% and 35.8%, respectively (p < .001). CCRT was an independent factor for better OS (HR, 0.69; 95% CI, 0.50-0.97; p = .034). CONCLUSION Pretreatment CEA levels in patients with SCC of the uterine cervix provide complementary information for predicting LF, DM, DFS, and OS, except for in patients with abnormal SCC-Ag levels before RT alone. More aggressive therapy might be advisable for patients with CEA levels of ≥10 ng/mL.

A Prospective Cohort Study to Compare Treatment Results Between 2 Fractionation Schedules of High-Dose-Rate Intracavitary Brachytherapy (HDR-ICBT) in Patients With Cervical Cancer

PURPOSE To compare the treatment results of 2 fractionation schedules for high-dose-rate intracavitary brachytherapy (HDR-ICBT) in patients with cervical cancer. METHODS AND MATERIALS From June 2001 through January 2008, 267 patients with stage IB-IVA cervical cancer were enrolled in the study. All patients underwent 4-field pelvic irradiation and HDR-ICBT. The median central and parametrial doses were 39.6 Gy and 45 Gy, respectively. Patient underwent either 6 Gy×4 (HDR-4) (n=144) or 4.5 Gy×6 (HDR-6) (n=123) to point A of ICBT using 192Ir isotope twice weekly. The rates of overall survival, locoregional failure, distant metastasis, proctitis, cystitis, and enterocolitis were compared between HDR-4 and HDR-6. RESULTS There were no significant differences in the demographic data between HDR-4 and HDR-6 except for total treatment time. The 5-year proctitis rates were 23.0% and 21.5% in HDR-4 and HDR-6 (P=.399), respectively. The corresponding rates of grade 2-4 proctitis were 18.7% and 9.6% (P=.060). The corresponding rates of grades 3-4 proctitis were 5.2% and 1.3% (P=.231). Subgroup analysis revealed that HDR-4 significantly increased grade 2-4 proctitis in patients aged≥62 years old (P=.012) but not in patients aged<62 years (P=.976). The rates of overall survival, locoregional failure, distant metastasis, cystitis, and enterocolitis were not significantly different between HDR-4 and HDR-6 schedules. CONCLUSION The small fraction size of HDR-ICBT is associated with grade 2 proctitis without compromise of prognosis in elderly patients. This schedule is suggested for patients who tolerate an additional 2 applications of HDR-ICBT.

3D Volumetric Visualization with Automatic Rigid and Deformable Hybrid Image Registration for Adaptive Radiotherapy

To provide more clinically convenient image fusions for adaptive radiotherapy (ART), an automatic rigid and deformable image registration framework (AIRF) is developed for multimodal visualization of multiple chronological images and multiple radiotherapy (RT) plans. Our hybrid image registration framework, AIRF, uses a faster but less accurate rigid registration method to provide an initial registration, followed by a slower but more accurate deformable registration method to fine tune the final registration. A multi-resolution approach is also employed in the image registration process to further improve the registration accuracy, robustness and efficiency. Volume visualization is provided to guide the automatic image registration process because it can reduce the global positioning error that results from a partial 3D visual presentation in the three conventional orthogonal planar views (axial, sagittal, and coronal). The AIRF can automatically align multiple volumetric images of patients taken over an extended period of time and can merge multiple radiotherapy plans based on different planning computed tomography (CT) images. It offers illustrative 3D volumetric visualization, hybrid rigid and deformable image registration, and automatic transfer of RT dose distribution and RT structure models such as treatment targets and organs at risk (OARs) onto chronological images. The AIRF can automatically register multiple volumetric image datasets of patients taken over an extended period of time and can merge multiple RT plans based on different planning CT images for 4D or adaptive radiotherapy.

Differences in the outcomes and complications between elderly and younger uterine cervical cancer patients treated by definitive radiotherapy — A propensity score-matched study

PURPOSE To evaluate the differences in the treatment outcomes and complications between elderly patients and younger patients with uterine cervical cancer (CxCa). METHODS AND MATERIALS From April 1993 to December 2007, 138 CxCa patients aged ≥75years (Elderly group) and 334 CxCa patients aged <60years (Young group) who underwent definitive radiotherapy/chemoradiotherapy at our institution were reviewed. Two propensity score-matched cohorts of patients were selected from both age groups to evaluate the differences in the outcomes and complications. The overall survival (OS), cancer-specific survival (CSS), local failure (LF), distant failure (DF), late proctitis, and cystitis were compared between the age groups. RESULTS The median follow-up time for survivors was 60.6months. A cohort of 99 pairs of patients was selected for the outcome comparison; there was a significant difference in the 5-year OS between the Elderly and Young groups (49.2% and 71.5%, respectively; p<0.001) but no differences in CSS, LF, and DF. Another cohort of 79 pairs of patients was selected for complication analysis. Significant differences between the Elderly and Young groups were observed in the 5-year cumulative grade 2 proctitis (39.7% and 17.2%, respectively; p=0.015) and grade 3 proctitis (18.1% and 6.2%, respectively; p=0.040). CONCLUSIONS Although OS was worse in the elderly patients, no differences were observed in CSS, LF, and DF. Meanwhile, elderly patients tended to have higher radiation-related proctitis than younger patients. A more conservative treatment strategy for elderly CxCa patients is reasonable in our future practice.

Quantitative Coronary Analysis Medical Image Processing Improved by Combining Wavelet Edge Detection and Segmentation

The Quantitative Coronary Analysis (QCA) is a useful method for physicians to diagnose heart artery stenosis. So the precision of coronary angiography image quality is very definite important. In this study we present a precision 2-D image processing method that combines gradient segmentation and region segmentation approaches with an entropy maximization procedure. This method allows us to utilize all available information to achieve the most robust segmentation results for coronary angiography analysis. The aim is to improve the methods of 2-D coronary angiography image processing those usually lack such as precision. The example of coronary angiography data of true medical images were used to test the validity of our method. We found that our method not only achieved the precision we sought but also has many interesting applications that shall be most useful in medical practices.