Objective To identify high-risk group among gastric cancer patients treated with curative resection and more than D1 dissection, and investigate the indications for proper adjuvant therapy.Methods 297 patients who met the following enrolled criteria were retrospectively analyzed:treated between January 2002 and December 2004, primary gastric or gastroesophageal cancer, underwent curative gastrectomy and more than D1 lymphadenectomy, pathologically staged as T3-4N0-1M0,or TxN2-3M0.The overall survival (OS), disease-free survival (DFS), local-regional recurrence-free survival (LRFS) and distant metastasis-free survival (DMFS) were calculated, and possible prognostic factors were analyzed.Results The median follow-up time was 61 months.The follow-up rate was 92.3%.The 5-year OS, DFS, LRFS and DMFS were 57.9%, 52.2%, 70.6% and 71.7%, respectively.Four independent prognostic variables identified for OS, DFS, LRFS and DMFS using multivariate analysis were Borrmann type (Ⅰ+Ⅱ/Ⅲ+Ⅳ), total number of dissected lymph nodes (>18/≤18), number of positive lymph nodes (0-3/≥4), and 6th AJCC TNM stage (Ⅱ+Ⅲ a/Ⅲ b+ⅣM0)(χ2=3.94-16.34,P<0.05).If one unfavorable prognostic factor was scored as 1, according to the total scores of the four prognostic factors, four risk groups were generated as low (score:0), low-intermediate (score:1), high-intermediate (score:2) and high risk group (score:3 or 4).The 5-year OS, DFS, LRFS and DMFS were 85.7%, 61.0%, 58.6% and 38.6%(χ2=31.20,P<0.01) in low risk group, 85.2%, 61.3%, 48.1% and 31.8%(χ2=31.88,P<0.01) in low-intermediate risk group, 94.4%, 77.8%, 64.4% and 57.2%(χ2=18.36,P<0.01) in high-intermediate risk group and 87.9%, 75.0%, 74.2% and 55.5%(χ2=19.30,P<0.01) in high risk group.Conclusions Even with R0 resection and more than D1 lymphadenectomy, the outcome was poor for gastric cancer patients with two or more unfavorable prognostic factors.Prospective study is warranted to evaluate the efficacy of adjuvant concurrent chemoradiotherapy for this group of patients.

Objective To evaluate the role of postmastectomy radiotherapy in four subgroups of high-risk breast cancer patients, who were grouped by the status of estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor 2 (Her-2). Methods A total of 437 invasive breast cancer patients with T3-4N1 or N2-3 and available immunohistochemistry results of ER, PR and Her-2 were retrospectively analyzed. Patients were classified into 4 subgroups according to hormone receptors (ER or PR, Rec) and Her-2 status:Rec-/Her-2-(triple negative), Rec-/Her-2 +, Rec +/Her-2 + and Rec +/Her-2-. Rec-was defined as ER-and PR-. Rec + was defined as ER + and/or PR +. Her-2 positive was defined as Her-2 + + or Her-2 + + +. End points were isolated locoregional recurrence (LRR), distant metastasis (DM), disease-free survival (DFS) and overall survival (OS). Results The median follow up time was 48 months. Sixty-nine (15. 8%) patients were Rec-/Her-2-, 62 (14. 2%) Rec-/Her-2 +, 89 (20.4%) Rec +/Her-2 + and 217 (49.7%) Rec +/Her-2-. 480(93.4%) patients received chemotherapy and 352(80. 5%) received radiotherapy. Radiotherapy significantly reduced the 5-year LRR rates of all the four subgroups (Rec-/Her-2-: 13.1% vs. 33. 3%, Rec-/Her-2 + :9. 3% vs. 21.2%, Rec + /Her-2 + :9. 7% vs. 47.0%, Rec +/Her-2-:3. 2% vs. 15.4%). Radiotherapy significantly lowered the 5-year DM rates (26. 7% vs. 49.4%, 27.6% vs. 67. 5%, 18.4% vs. 100%) and improved the 5-year DFS rate (66. 7% vs. 33. 3% , 67.7% vs. 33. 3% , 72. 6% vs. 0%) as well as OS (73.9% vs. 25.2% ,69. 8% vs.41.5%, 91.0% vs. 32. 8%) of patients with Rec-/Her-2-, Rec-/Her-2 + and Rec +/Her-2 +. Conclusions In high-risk breast cancer patients, all subgroups of patients grouped by ER, PR and Her-2 status can benefit from postmastectomy radiotherapy.

ObjectiveTo evaluate the inter-and intra-fractional clinical target volume (CTV) positioning errors of patients receiving postoperative simplified intensity-modulated radiotherapy (SIMRT) using cone beam computed tomography (CBCT).MethodsTwelve patients with liver tumors underwent postoperative SIMRT.CBCT images were acquired before and after the treatment.The clipbox volume for registration included the fiducial markers in the tumor bed and excluded the ribs and vertebral bodies.If any translational parameter of setup error before treatment exceeded 3 mm or rotational parameter exceeded 3°,the treatment couch was adjusted and a verification CBCT was acquired to assess residual setup error.Automatic bone match was used.A total of 214 acquisitions of CBCTs in 111 groups were analyzed.Inter-fractional translational CTV positioning errors in left-right (x),superior-inferior (y) and anterior-posterior (z) axis were calculated in 111 groups,and intra-fractional translational CTV positioning errors in 70 groups.Clinical to planning target volume (PTV) margins were calculated according to the formula:margin =2.0 ∑ + 0.7σ ( ∑ is systematic error,σ is random error).ResultsInter-fractional translational CTV positioning errors in x,y and z axis were -0.03 mm,-0.43 mm,1.02 mm,with systematic error ( ∑ ) of 1.50 mm,5.89 mm,1.97 mm,and random error (σ) of 1.76 mm,4.13 mm,2.42 mm,respectively.Intra-fractional translational CTV positioning errors in the x,y,z axis were 0.04 mm,0.86 mm,- 0.46mm,with systematic error (∑) of 0.46 mm,1.14 mm,0.31 mm,and random error (σ) of 0.95 mm,1.38 mm,0.91 mm,respectively.The calculate CTV to PTV margins were 4.5 mm,15.0 mm,5.8 mm in the x,y,z axis,respectively.ConclusionsThe CTV errors were inevitable when patients with liver tumors received SIMRT.Fiducial markers placed in tumor bed during operation were helpful for accurate positioning error analysis.

Purpose To investigate MRI features of encephalitis in children with severe hand-foot-mouth disease (HFMD). Materials and Methods The MR images of 85 cases of clinically diagnosed encephalitis in severe HFMD patients were retrospectively reviewed. Results In 85 patients, plain MRI was negative in 28;57 cases were positive including 10 cases showing punctate or patchy lesions in cerebral hemisphere, callosum, thalamus and cerebellar vermis with isointensity or slightly hypointensity on T1WI, isointensity or slightly hyperintensity on T2WI, and hyperintensity on T2FLAIR and DWI. In 47 cases, the lesions were located in the brainstem, tegmentum of pons and medulla oblongata. MRI features included symmetric high signal on DWI in posterior portions of brainstem, hypointensity or isointensity on T1WI and hyperintensity or isointensity on T2WI and T2FLAIR (n=17);patchy long T1, long T2 signal in posterior portions of brainstem, normal or mild hyperintensity on DWI, hyperintensity on T2FLAIR (n=30); lesion extending around the fourth ventricle in 9 patients, one of which involved bilateral cerebral peduncles. Conclusion Encephalitis in children with severe HFMD has characteristic MRI features. The lesions are mostly located in the posterior portion of the brainstem. DWI and T2FLAIR can better detection lesions.

Objective To evaluate the risk of locoregional recurrence ( LRR ) and role of radiotherapy for patients with estrogen receptor?negative and human epidermal growth factor receptor 2?overexpressed ( Rec?/HER?2+) locally advanced breast cancer ( LABC ) . Methods A retrospective analysis was performed on the clinical data of 294 patients with Rec?/HER?2+LABC from 1999 to 2011. All patients were treated with modified radical mastectomy ( MRM ) . Of them, 239 patients received postmastectomy radiotherapy and 55 patients did not. Locoregional recurrence?free survival ( LRRFS) and overall survival ( OS) , as well as LRR, were compared between the two groups. The Kaplan?Meier method was used to estimate survival and recurrence rates, and the log?rank test was used for survival difference analysis and univariate prognostic analysis. Multivariate prognostic analysis was performed using the Cox regression model. Results The 5?year sample size was 162. Fifty?six patients developed LRR. The 5?year LRRFS and OS rates were 79. 7% and 70. 0%, respectively. Postmastectomy radiotherapy significantly increased the 5?year LRRFS rate ( 85. 1% vs. 56. 0%, P=0. 000) , but did not significantly increase the 5?year OS rate ( 71. 3% vs. 64. 2%, P= 0. 441 ) . Multivariate analysis indicated that postmastectomy radiotherapy was the only independent prognostic factor associated with increased LRRFS ( RR=0. 303, 95% CI:0. 166?0. 554, P=0. 000). Conclusions Patients with Rec?/HER?2+ LABC treated with MRM alone appear to be at a significantly increased risk of LRR compared with those treated with MRM followed by radiotherapy.

Objective To assess the prognostic benefits of intraoperative radiotherapy (IORT) with electron beam among patients with unresectable locally advanced pancreatic cancer.Methods Between January 2009 and December 2014,167 patients with unresectable locally advanced pancreatic cancer received IORT with electron beam (10-20 Gy) in our hospital.After surgery,12 patients were treated with external beam radiotherapy,56 patients with chemoradiotherapy (CRT),and 17 patients with chemotherapy.Overall survival (OS),local recurrence,and toxicities were retrospectively analyzed.The Kaplan-Meier method was used to calculate survival rates,the log-rank test was used for survival difference analysis and univariate prognostic analysis,and the Cox model was used for multivariate prognostic analysis.Results The follow-up rate was 100％.The median OS time was 10.3 months,and the 2-year OS rate was 22％.The median progression-fiee survival (PFS) time was 6.3 months,and the 2-year PFS rate was 9.9％.The cancer-specific survival (CSS) time was 11.2 months,and the 2-year CSS rate was 23.6％.In the patients treated with IORT alone at doses of＜15 Gy,15 Gy and＞15 Gy,the median OS times were 6.2 months vs.9.1 months vs.22.2 months,and the 1-year OS rates were 10.0％ vs.39.6％ vs.74.4％ (P=0.000).Among the patients receiving postoperative adjuvant therapy,those treated with IORT+CRT had the best survival,with a median OS time of 11.6 months (P=0.033).The univariate analysis showed that IORT dose (P =0.000),tumor size (P =0.006),and IORT applicator diameter (P =0.007) were prognostic factors.The multivariate analysis showed that IORT dose (P=0.000) and IORT combined with CRT (P=0.006) were independent prognostic factors.Conclusions IORT with electron beam is an effective and safe treatment strategy for unresectable locally advanced pancreatic cancer.After protecting surrounding organs,increasing the IORT dose can improve the survival.IORT combined with CRT should be recommended because it improves survival for unresectable locally advanced pancreatic cancer without increasing toxicities.

ObjectiveTo compare the role of AJCC 6th and 7th TNM staging systems in predicting the long term survival of locally advanced gastric cancer patients after curative surgery. Methods All patients who met the following criteria were included for analyses: treated between January 2002 and December 2004, primary gastric or gastroesophageal cancer, underwent curative gastrectomy ( UICC R0 ) and at least more than D1 lymphadenectomy, pathologically staged as T3-4N0-1 M0, or any T, N2-3M0. Overall survival (OS), disease-free survival ( DFS), locoregional recurrence-free survival (LRFS) and distant metastasis-free survival (DMFS) were calculated and compared according to N and TNM stage of the two TNM staging systems (T stage were not analyzed due to its extremely unbalanced distribution). The N and TNM stage and other significant variables in univariate analyses were evaluated further for both OS and DFS by Cox regression. ResultsThe median follow-up time was 61 months. The follow-up rate was 92. 3％.Among the 297 enrolled patients, 56. 9％ of the patients had a stage migration between the two staging systems. According to 6th, no difference in DMFS was detected between different N and TNM stages (x2 =6. 65, P =0. 084 ; x2 =6. 61, P =0. 108 ). When using 7th, statistically significant difference was found in DMFS between different N stages ( x2 =9. 035,P =0. 029), and TNM stage also tended to have an influence on DMFS ( x2 =7.27,P =0. 064). The N and TNM stage had similar significant influence on OS, DFS and LRFS based on both staging systems ( x2 =9. 23 - 19. 00,P =0. 000 -0. 026 and x2 =11.67 - 19. 11 ,P =0. 000 -0. 009).In Cox regression, TNM stage was an independent prognostic variable for both OS and DFS based on these two staging systems (x2=9.05-25.51, P=0.000-0.003 ), but the 7thappeared to be a better predictor than the 6 th ( OS : RR =1.6 1 8 vs 1.4 9 6 ; DFS : RR =1.5 9 4 vs 1. 5 6 4 ).ConclusionsThe N and TNM stage in 7th TNM staging system are more predictive for DMFS than in 6th TNM staging system for locally advanced gastric cancer patients. The 7th TNM staging system provides a better prognostic estimation of both OS and DFS.

Objective The benefit of adjuvant chemoradiotherapy remains controversial for gastric cancer patients treated with more than D1 dissection. This retrospective analysis is to distinguish the first site of recurrence in patients treated with curative resection and more than D1 dissection and to find any feasible adjuvant concurrent chemoradiotherapy recommendation for them. Methods All patients treated between January 2002 and December 2004 who met the following criteria were analyzed: primary gastric or gastroesophageal cancer, underwent curative gastrectomy ( UICC R0 ) and more than D1 lymphadenectomy,pathologically staged as T3-4N0-1 M0, or any Tx N2-3M0. There were 297 patients analyzed and 19.5%,52. 2%, 17. 8% , 10. 4% of patients had stage Ⅱ ( T3 N0 M0, T1 N2 M0 ), Ⅲa, Ⅲb and Ⅳ ( M0 ) diseases,respectively. 76. 1% of patients received adjuvant chemotherapy, while Only 2 patients underwent adjuvant radiotherapy. Failure patterns and the prognostic factors for locoregional recurrence were analyzed. Results The median follow-up time was 61 months and the follow-up rate was 92. 3%. 145 patients developed recurrence with a median recurrent time of 26 months. Locoregional recurrence was observed in 82 patients and distant metastasis in 79 patients. Gastric stump, anastomosis, intra-abdominal lymph nodes were the most common sites of locoregional recurrence. Liver and lung were the most frequent sites of distant metastasis. Prognostic variables for locoregional recurrence were identified after univariate analysis,including pathologic type ( χ2 = 11.50, P = 0. 009 ), total number of dissected lymph nodes ( χ2 = 6. 65,P =0. 010), the number of positive lymph node ( χ2 =5. 80,P =0. 016), lymph node capsular invasion ( χ2 =pathologic type, total number of dissected lymph nodes, lymph node capsular invation, AJCC TNM stage and Borrmann type were independent prognostic factors for locoregional recurrence ( χ2 = 6. 77,19. 33,17. 84 and 6. 02,P =0. 009,0. 000,0. 000 and 0. 014). ConclusionsLocoregional recurrence remains the main cause of failure for locally advanced gastric or gastroesophageal cancer patients even though the patients have had more than D1 lymphadenectomy. The role of adjuvant concurrent hemoradiotheray for those patients is warranted.

Objective To analyze the rule of lymph node metastasis in thoracic esophageal carcinoma,and to study the proper radiation target. Methods From September 1986 to December 1997,549 patients with esophageal carcinoma who had undergone radical reseetion were divided into surgery alone group (S,275 patients) or surgery plus radiotherapy group( S + R,274 patients). Radiotherapy was begun 3 to 4 weeks after operation. The radiation target included beth supra-clavicular areas and the entire mediastinum. The total dose was 50 Gy in 25 fractions over 5 weeks for the supra-clavicular areas and 60 Gy in 30 fractions over 6 weeks for the entire mediastinum. Results The 5-year overall survival of patients with lymph node metastasis in one anatomic site and two anatomic sites was 31.5% and 13.9% (P=0.013), respectively. For patients with > 2 positive nodes metastasis receiving surgery alone, the corresponding 5-year survival was 24.8% and 4.9% (P=0.046) ,respectively. The median number of dissected lymph nodes of the upper-,middle-and lower-segment esophageal carcinoma was 13,17 and 20, respectively. The rate of metastatic lymph node in the para-esophagus region was the highest(61.5%-64.9%) ,which was not different among the different primary sites(P=0.922). The anastomotic stoma recurrence rate of the upper-segment esophageal carcinoma was higher than that of the middle- or lower-segment carcinomas(16.7% ,3.1% ,and 7.7%, χ2=9.02,P<0.05). Conclusions For the thoracic esophageal carcinoma,the number of anatomic sites of lymph node metastasis is an important factor affecting the survival. The lower rate of lymph node metastasis of the upper segment esophageal carcinoma may be corrected with the less lymph node dissected. The rate of lymph node metastasis in para-esophageal region is not related with the lesion segment. The anastemotie stoma is an important radiotherapy target for upper segment esophageal carcinoma.

Objective To investigate the treatment outcome and failure in patients with primary mediastinal large B-cell lymphoma(PMBL). Methods Between Jan. 1992 and Oct. 2006, a total of 46 patients with pathologically confirmed PMBL were reviewed, including 14 with Ann Arbor Stage I disease, 23 with Stage Ⅱ disease,3 with Stage Ⅲ disease and 6 with Stage Ⅳ disease. Stage Ⅰ+Ⅱ disease was present in 80% of the patients. All patients were treated with chemotherapy ,and 29 also received radiotherapy. Twenty-seven patients(59%) were treated with first generation regimen(CHOP),9(20%) with third generation regimens (MACOP-B, ProMACE/CytaBOM, m-BACOD, or ProMACE-MOPP), and 10(22%) with high-dose chemotherapy (HDCT/APBSCT). Rituximab was administered to 16 patients (35%). For most patients who received radiotherapy,an involved field was used with a median dose of 45 Gy in 23 fractions.Results The rate of complete remission, partial remission and progression disease was 41%, 30% and 24% ,respectively. The 5-year overall survival rate(OS) for all patients was 35%. The 2- and 5-year OS was 79% and 63% for stage Ⅰ+Ⅱ and 51% and 0 for stage Ⅲ+Ⅳ ,respectively(X2=4.35 ,P=0. 037).The 2-year progression free survival rate was 63 % and 11%, respectively (X2=17.77, P=0.1300). The 5-year OS was 80% for the patients with CR,50% for those with PR,and 0 for those with progression disease(X2= 19.58 ,P=0.003 ). With a median follow-up of 22 months, progression disease and relapse occurred in 19 patients. Conclusions Survival of patients with advanced stage PMBL is poor. Further studies areneeded to confirm the optimal treatment. Radiotherapy often plays a pivotal role in local control.