Li-Fraumeni Syndrome (LFS) is a rare cancer-predisposing disease caused by germline mutations in a potent tumor suppressor gene calledTP53. Accordingly, LFS patients carry a high lifetime risk of developing multiple primary cancers. Conventional wisdom and prior work have suggested that the risk of developing a subsequent malignancy is increased by the use of radiation therapy (RT) during primary or secondary cancer treatment. This risk, however, is not well characterized. Here we describe the risk of subsequent malignancy and cancer-related death in LFS patients after undergoing RT for a first or second primary cancer. We reviewed a cohort of patients in a multi-institution hereditary cancer registry with germline TP53 mutations. We assessed the rate of subsequent malignancy and death in patients who received RT (RT group) as part of their cancer treatment against those who did not (non-RT group). We further evaluated outcomes after RT with respect to p53 mutation type. Among 41 LFS patients (14 males and 28 females from 28 different families), 22 TP53mutations were present. 61% (25) and 7% (3) of patients had missense mutations in the DNA binding domain and tetramerization domain of p53, respectively, and 27% (11) had truncating mutations caused by nonsense (3), frameshift (1), splice site (2), or deletion (5) mutations. The median age of first cancer diagnosis was 22 years. 13 patients received RT with curative intent as part of their cancer treatment which included 3 for locally recurrent disease. The median time to follow up after RT was 5.0 years. 61% of patients in the RT group compared to 40% of patients in the non-RT group developed a subsequent malignancy (P=0.31). The median time to subsequent malignancy in the RT group and non-RT group was 3.2 years and 4.3 years respectively. Although 4 patients developed a subsequent malignancy within the radiation field, all were of the same initial histology and none of the cancers had features of an RT-induced secondary malignancy. To date, 8 deaths have occurred in the RT group (median age 36.5 years) and 2 deaths in the non-RT group (median age 30.5 years). The 5-year overall survival (OS) for the RT group was 54% compared to 95% in the non-RT group (P=0.033). Notably, there was no association between the type of p53 mutation and the rate of subsequent malignancy after RT. LFS patients who received RT for a primary cancer suffered from subsequent malignancies and death at a higher rate than those who did not receive RT. However, these results must be interpreted with caution as the majority of patients within the RT-treated group had more aggressive cancers with a higher risk of recurrence. We found that all 4 in-field malignancies in the RT-group likely represent locally recurrent disease and not RT-induced malignancies. Accordingly, the higher rate of mortality in the RT-group is likely a reflection of more aggressive cancer diagnoses and not their treatment.

Li-Fraumeni Syndrome (LFS) is a rare cancer-predisposing disease caused by germline mutations in a potent tumor suppressor gene calledTP53. Breast cancer is the most commonly diagnosed malignancy in women with LFS. The current treatment recommendations for LFS breast cancer patients includes bilateral prophylactic mastectomy with or without chemotherapy. Meta-analyses have shown that post-mastectomy radiation therapy (RT) can reduce the risk of recurrence and cancer-related mortality in patients with node positive disease. However, the risks and benefits of RT in the LFS patient population have not been well-described and there is significant concern of RT-induced secondary malignancies. Here we describe the incidence of cancer events in breast cancer patients with LFS treated with or without RT, including local regional recurrences and any in-field secondary malignancies, with the aim to explore whether the benefits of RT outweigh the risk of secondary malignancy associated with radiation exposure. We reviewed a cohort of patients with germline TP53mutations and the diagnosis of breast cancer in a multi-institutional hereditary cancer registry from 2004-2017. We assessed outcomes in regards to disease recurrence and subsequent malignancy. Of 41 LFS patients in our database, 11 had a first primary breast cancer diagnosis. The median age at diagnosis was 35. Six patients underwent bilateral mastectomy, 3 patients underwent unilateral mastectomy and 2 patients underwent lumpectomy without RT. Five patients received adjuvant chemotherapy and 3 patients received post-mastectomy RT. The median follow up time from the initial cancer diagnosis was 4.75 years. Among the 8 patients who did not receive RT, 1 developed ipsilateral recurrence in the chest wall 3.75 years after initial diagnosis, 2 developed contralateral breast cancer after 1.2 and 2.5 years, and 1 developed a new primary cancer after 21.5 years. Among the 3 patients who received RT, only 1 developed a secondary malignancy, a B-cell ALL after 3.3 years. There were no in-field secondary malignancies. Moreover, 1 patient received RT for local and regional disease recurrence after undergoing bilateral mastectomy and remains disease-free and without a secondary malignancy 12.0 years after completion of RT. Given the rarity of LFS, the literature on patients receiving RT for breast cancer is limited to small case series like our study. One series evaluated 8 breast cancer patients with LFS including 6 undergoing RT and found more cancer events in patients treated with RT, concluding that RT should be avoided. In contrast, we did not find any in-field secondary malignancies among our cohort. Although this study is purely observational, our results differ from prior studies and question whether RT should always be avoided in LFS patients. It is possible that in LFS patients with node-positive breast cancer, the benefits of RT could outweigh the risks of an RT-induced secondary malignancy.

Circulating lymphocytes are highly sensitive to radiation. Severe lymphopenia (grade 4) is associated with worse clinical outcomes in cancer patients. We sought to identify the association between cardiac dose and lymphopenia grade in non-small cell lung cancer (NSCLC) patients. A total of 138 patients with stage IIb-IV NSCLC were included in this study. They were among those enrolled in a randomized protons vs. IMRT concurrent chemoradiation therapy (CCRT) lung trial. Cardiac regions of interest (ROIs) were automatically delineated using an in-house developed multi-atlas contouring system, followed by manual editing by a thoracic radiation oncologist using the RTOG 1106 atlas contouring guidelines. The cardiac ROIs included pericardium, whole heart muscle, left and right ventricles, left and right atria, and large vessels. The dose-volume parameters was extracted for data analysis. Patient's complete blood count data were collected, which included baseline, at least weekly during CCRT, and on each follow-up. The absolute lymphocyte counts (ALCs) were extracted and lymphopenia was graded according to CTCAE v 3.0. The equality of group medians was assessed with a nonparametric test with p≤0.05 indicating significance. There were 95 IMRT and 43 PSPT patients. Tumor location were left upper lobe 31, left lower lobe 15, mediastinum 5, right upper lobe 60, right middle lobe 7, right lower lobe 18, and unknown primary tumor location in 2 (Metastatic mediastinal lymph node carcinoma). The worst lymphopenia grade was: 0-2 in 13 (9.4%) patients, grade 3 in 70 (50.7%) patients, and grade 4 in 55 (39.9%) patients. All the cardiac dosimetric parameters were highly correlated with the worst lymphopenia grade. Table 1 shows the relationship between dose-volume indices of some of the cardiac substructures and the worst lymphopenia grade (0-2 vs. 3 vs. 4). Results from this study show that percentages of V5 to V10 for cardiac muscle, pericardium, heart chambers, and descending aorta are significantly associated with the worst grade lymphopenia in locally advanced NSCLC.Tabled 1Abstract TU_35_3667; Table 1The worst lymphopenia grade0-234Median Test p-valueThe worst lymphopenia grade0-234Median Test p-valueCases137055137055Cardiac Muscle V5 (%)223952<0.001Cardiac Muscle V10 (%)173143<0.001Pericardium V5 (%)365060<0.001Pericardium V10 (%)3143520.001Left Ventricle V5 (%)622320.002Left Ventricle V10 (%)313200.001Right Ventricle V5 (%)1228450.005Right Ventricle V10 (%)721370.002Left Atrium V5 (%)3761750.001Left Atrium V10 (%)2951650.001Right Atrium V5 (%)184063<0.001Right Atrium V10 (%)143256<0.001Descending Aorta V5 (%)3859630.001Descending Aorta V10 (%)3250510.006 Open table in a new tab

Bolus cisplatin (BC [100 mg/m2]) concurrent with head and neck radiation therapy (RT) is associated with high rates of toxicity and a poor completion rate. Weekly cisplatin (WC [40 mg/m2]) has been used in our institution since 2003. We present the rates of radiation and chemotherapy treatment completion, elapsed radiotherapy time, overall survival (OS), disease free survival (DFS), and distant metastasis free survival (DMFS) using WC, BC, or cetuximab. We reviewed 336 consecutive cases of head and neck cancer treated with concurrent WC x 6 cycles (n=234), BC x 3 cycles (n=50), or cetuximab with 1 induction and 6 weekly concurrent cycles (n=52) from 2003 to 2015. Definitive RT was prescribed to 66-70 Gy in 2.0-2.25 Gy fractions (median 67.5 Gy in 2.25 Gy fractions) and adjuvant RT was prescribed to 60 to 67.5 Gy in 2.0-2.25 Gy fractions (median 66 Gy in 2.0 Gy fractions). IMRT was generally used except for the oral tongue. Treatment stoppage and/or breaks were triggered by Stage III or IV acute toxicity or patient wishes. Chi-square was used to determine differences across groups and Kaplan-Meyer with the log rank test was used in survival analyses. Three hundred thirty-six cases were analyzed with a mean follow up of 31 months. Primary site was oropharynx in 211, oral cavity in 47, larynx in 46, hypopharynx in 14, nasopharynx in 12, and unknown primary in 6. HPV was positive in 75 of 89 (84%) oropharynx cases where HPV status was known. AJCC staging was stage I in 2, stage II in 10, stage III in 36, and Stage IVA/B in 288. Histology was squamous cell carcinoma in 329/336 patients. Stage, primary site, and HPV status were balanced across treatment groups. The full prescribed course of radiation was completed in 224/234 (96%) patients receiving WC, 48/50 (96%) of patients receiving BC, and 44/52 (85%) patients receiving cetuximab (p=0.01). The full course of concurrent chemotherapy was completed in 205/234 (88%) patients receiving WC, 7/50 (14%) patients receiving BC (44 [88%] completed at least 2 cycles), and 42/52 (81%) patients receiving cetuximab (p<0.01). The mean cumulative dose of cisplatin given was 221 mg/m2 for patients receiving weekly cisplatin and 204 mg/m2 for patients receiving bolus cisplatin (p=0.62). There was no difference in mean elapsed treatment time between groups. For patients treated definitively with concurrent WC, BC, and cetuximab, median OS was 121 mo., 76 mo., and 51 mo. respectively (p=0.04 for weekly cisplatin vs. cetuximab, NS for other comparisons); and median DFS was 121 mo., 75 mo., and 38 mo. respectively (p=0.02 for WC vs. cetuximab, NS for other comparisons). At 3-years, distant metastases had developed in 12/324 (4%) patients receiving WC, 4/50 (8%) patients receiving BC, and 2/52 (4%) patients receiving cetuximab with no differences in DMFS. Concurrent WC with head and neck irradiation is associated high rates of chemotherapy and RT completion and compares favorably to BC and cetuximab in OS, DFS, and DMFS.

To evaluate long-term outcomes of HB-IMRT/AP in patients with advanced head and neck cancer (HNC) with regard to radiation dose to the larynx, cervical esophagus and cervical segmental spinal cord and locoregional control. From 2/2003 through 5/2016, 99 consecutive patients with HNC treated with definitive RT or chemoRT, were included in the study cohort (7 nasopharynx, 2 oral cavity, 89 oropharyngeal, 1 unknown primary.) Exclusion criteria included patients with either primary or neck surgery, primary cancer in the larynx or hypopharynx, ipsilateral neck RT, or inferior border of primary tumor to the isocenter match line was less than 1.5 cm. Nodal disease across the isocenter match-line was allowed with additional in-field AP boost to achieve full coverage of prescription dose. There were 94 males and 5 females with mean age 60.8 years old (range, 40-85); 94 of 99 (95%) of patients were stage III-IVB, 5 stage II. Ninety-seven patients received concurrent chemoRT, with 75.8% receiving weekly cisplatin at 40mg/m2, median of 6 cycles (range, 2-6). Simultaneous in-field boost technique was used for HB-IMRT with median dose 67.5 Gy (65.25-75 Gy) to the upper neck GTVs, and 50 Gy (44-54 Gy) to lower neck AP field, 55 patients received small AP field(s) boost 10-17.5 Gy to either nodal disease across the match line or inadequate margins from the upper neck nodal sites. On AP field, the larynx, upper cervical esophagus and spinal cord were blocked. For patients with very low nodal risk, the entire AP field midline was blocked to maximize sparing of normal structures. Match-line failure was defined as recurrence within 1.5 cm of the match line. With HB-IMRT/AP technique, mean laryngeal dose was 14.5±7 Gy; mean upper cervical esophagus dose 12.5±4.6 Gy with block vs. 43.9±5.5 Gy without block (p<0.001); mean upper cervical spinal cord dose 10.7±4.0 Gy with block versus 35.7±4.8 Gy without block (p<0.001). No patient became long-term PEG tube dependent or required prolonged feeding tube use. With median follow-up of 31.4 months, there was no nodal recurrence, 7 primary tumor site failures, and 7 distant metastases. Two- and 5-year LC, DFS, DSS and OS were 91.9% (95% CI: 86-97.7%) and 89.4% (CI: 81.9-96.9%), 88.7% (95% CI: 80-93.8%) and 84.5% (95% CI: 73.8-91.2%), 94.3% (95% CI: 86.8-97.6%) and 84.5% (95% CI: 69.5-92.4%), 87.2(95% CI: 78-92.7%) and 72.85(95% CI: 58.6-82.9%), respectively. There were no match-line failures. All 7 local failures occurred above the match-line with mean 5.3 cm distance (2.5-7.0 cm) from the lowest border of recurrence to the match-line. HB-IMRT/AP for advanced HNC significantly reduced dose to critical structures without compromising locoregional control. Careful selection of the patients for HB-IMRT/AP is important to avoid inadequate margins in the primary sites.

Limited data have been published regarding the effect of adaptive radiotherapy (ART) on clinical outcome in patients with nasopharyngeal carcinoma (NPC). We compared the long-term outcomes in patients with locally advanced NPC treated by adaptive intensity-modulated radiotherapy (IMRT) replanning versus IMRT.

To preliminary analyze the treatment response and radiation induced lung injury (RILI) of metastatic non-small cell lung carcinoma (NSCLC) patients treated with a novel hypo-IMRT radiation modality. Between January 1, 2006, and December 31, 2013, metastatic NSCLC patients were enrolled in our prospective study. The diameter of the tumor was larger than 3 cm. All the patients received adaptive radiation therapy (ART) hypo-IMRT radiation modality. For some special patients, dose painting (DP) was used. GTV was outlined based on the visible tumor lesions of 4D-CT, and the expansion of 3mm was outlined as PTV. For dose painting patients, GTV1 was the central region of the visible tumor lesion. GTV2, GTV3, and GTV4 were the lesions near the esophagus, heart, and lung respectively. The radiation dose ranged from 2.5Gy-20Gy/f. The total biological effective dose (BED) (α/β=10) was greater than 70Gy. All the patients received radiation once a day. According to RTOG0617 recommendation, the constraints for lung were V20≤37%, V5≤65%. Platinum-based chemotherapy regimens were given for 2-6 cycles. Acute radiation induced pneumonitis (ARP) and radiation induced pulmonary fibrosis (RPF) were evaluated within and after 6 months respectively. SPSS was used to analyze the correlation between treatment response and incidence of RILI. 39 metastatic lung cancer patients (51 lesions) were included in the study. 35 patients used ART, the median physical dose was D90 54.3Gy (45-96.35 Gy), the fraction number was 8.5 (3-26). Median BED for GTV (α/β=10) was 100.8Gy (84-160Gy). 4 patients used the does painting, the D90 of physical dose was 92Gy (69-119Gy), 62Gy (57.5-69Gy) for GTV1 and GTV2-4 respectively. The D90 of BED (α/β=10) was 120Gy (89.7-180Gy) and 82Gy (72-90Gy) for GTV1, GTV2-4. The median followed-up time was 16 months. CR and PR was 65.0%. Survival analysis showed that 6 months, 1 yrs, 2 yrs OS and LC were 97.5%, 85.6%, 47.8% and 96.7%, 96.7%, 80.6%, respectively. LC for metastatic lesions was 98.1%, 82.3%, 69.5% respectively. The median lesions progression-free time was 40.9 months. Univariate analysis showed that COPD (p=0.030), ART+DP (P=0.008) expressed the benefit on LC. Multivariate analysis showed the factors above expressed no benefit on LC. Univariate and multivariate analysis showed no factors expressed benefit on OS. The incidence of APR was 71.8% and 28.2% for ≤G2 and ≥G2. The incidence of RPF was 39.4% and 10.6% for ≤G2 and ≥G2. There was no significant correlation between radiation dose and radiation induced lung injury for metastatic lung cancer patients. The treatment response and radiation induced lung injury of metastatic NSCLC patients treated with a novel hypo-IMRT radiation modality were acceptable in clinic. Dose painting might improve the local control of patients with MLC. There was no significant difference between radiation dose and induced lung injury.

Purpose:To evaluate the feasibility of using an automatic segmentation tool to delineate cardiac substructures from computed tomography (CT) images for cardiac toxicity analysis for non‐small cell lung cancer (NSCLC) patients after radiotherapy.Methods:A multi‐atlas segmentation tool developed in‐house was used to delineate eleven cardiac substructures including the whole heart, four heart chambers, and six greater vessels automatically from the averaged 4DCT planning images for 49 NSCLC patients. The automatic segmented contours were edited appropriately by two experienced radiation oncologists. The modified contours were compared with the auto‐segmented contours using Dice similarity coefficient (DSC) and mean surface distance (MSD) to evaluate how much modification was needed. In addition, the dose volume histogram (DVH) of the modified contours were compared with that of the auto‐segmented contours to evaluate the dosimetric difference between modified and auto‐segmented contours.Results:Of the eleven structures, the averaged DSC values ranged from 0.73 ± 0.08 to 0.95 ± 0.04 and the averaged MSD values ranged from 1.3 ± 0.6 mm to 2.9 ± 5.1mm for the 49 patients. Overall, the modification is small. The pulmonary vein (PV) and the inferior vena cava required the most modifications. The V30 (volume receiving 30 Gy or above) for the whole heart and the mean dose to the whole heart and four heart chambers did not show statistically significant difference between modified and auto‐segmented contours. The maximum dose to the greater vessels did not show statistically significant difference except for the PV.Conclusion:The automatic segmentation of the cardiac substructures did not require substantial modification. The dosimetric evaluation showed no statistically significant difference between auto‐segmented and modified contours except for the PV, which suggests that auto‐segmented contours for the cardiac dose response study are feasible in the clinical practice with a minor modification to the PV vessel.

Purpose/Objective(s)Neoadjuvant was a promising chemotherapy modality for head and neck cancer. However, there was still controversy surrounding it. We conducted a retrospective analysis to evaluate the early survival results of locally advanced nasopharyngeal carcinoma (NPC) patients treated with neoadjuvant chemotherapy and aim to explore the substantial prognostic factors further.Materials/MethodsOne hundred forty-seven squamous NPC patients of stage III-IVa+b (UICC2002 stating system) treated with intensity modulated radiation therapy (IMRT) in our center from October 2007 to October 2013 were analyzed retrospectively. They were divided into 2 groups. Group NACT (76 patients) was treated with neoadjuvant chemotherapy followed with concurrent chemoradiation therapy, and group CCRT (71 patients) was treated with concurrent chemoradiation therapy. NACT received 2 cycles of neoadjuvant chemotherapy with the TP (docetaxel 75 mg/m2 d1 + cisplatin 80 mg/m2 d1) or TPF (docetaxel 75 mg/m2 d1 + cisplatin 80 mg/m2 d1 + 5-Fu 750 mg/m2d1-5) regimens, and then received 2 to 3 cycles of the platinum-based concurrent chemoradiation therapy. CCRT only received 2 to 3 cycles of concurrent chemoradiation therapy. Cumulative survival rate was analyzed by Kaplan Meier method. The log-rank test and Cox proportional hazard model was used for prognostic analyses.ResultsThe median follow-up time was 26 and 30 months in the NACT and CCRT groups, respectively. The 3-year LRRFS, DMFS, and OS for groups NACT and CCRT were 94.5%, 96.8%; 85.8%, 81.6% and 83.3%, 85.0%, respectively (P<.05). Age, sex, HGB, clinical stage, T stage, N stage, and radiation dose were included in prognostic analysis. For univariate analysis, only N stage was the adverse prognostic factor for 3-year DMFS (P<.05). However, for multivariate analyses, there was no statistical significance for these possible prognostic factors.ConclusionThere was no statistical significance for early survival results in NACT and CCRT groups; however, the neoadjuvant chemotherapy followed with concurrent chemoradiation therapy could increase the DMFS for locally advanced NPC patients to some degree. N stage was also a possible prognostic factor for the survival results of the local advanced NPC patients. Purpose/Objective(s)Neoadjuvant was a promising chemotherapy modality for head and neck cancer. However, there was still controversy surrounding it. We conducted a retrospective analysis to evaluate the early survival results of locally advanced nasopharyngeal carcinoma (NPC) patients treated with neoadjuvant chemotherapy and aim to explore the substantial prognostic factors further. Neoadjuvant was a promising chemotherapy modality for head and neck cancer. However, there was still controversy surrounding it. We conducted a retrospective analysis to evaluate the early survival results of locally advanced nasopharyngeal carcinoma (NPC) patients treated with neoadjuvant chemotherapy and aim to explore the substantial prognostic factors further. Materials/MethodsOne hundred forty-seven squamous NPC patients of stage III-IVa+b (UICC2002 stating system) treated with intensity modulated radiation therapy (IMRT) in our center from October 2007 to October 2013 were analyzed retrospectively. They were divided into 2 groups. Group NACT (76 patients) was treated with neoadjuvant chemotherapy followed with concurrent chemoradiation therapy, and group CCRT (71 patients) was treated with concurrent chemoradiation therapy. NACT received 2 cycles of neoadjuvant chemotherapy with the TP (docetaxel 75 mg/m2 d1 + cisplatin 80 mg/m2 d1) or TPF (docetaxel 75 mg/m2 d1 + cisplatin 80 mg/m2 d1 + 5-Fu 750 mg/m2d1-5) regimens, and then received 2 to 3 cycles of the platinum-based concurrent chemoradiation therapy. CCRT only received 2 to 3 cycles of concurrent chemoradiation therapy. Cumulative survival rate was analyzed by Kaplan Meier method. The log-rank test and Cox proportional hazard model was used for prognostic analyses. One hundred forty-seven squamous NPC patients of stage III-IVa+b (UICC2002 stating system) treated with intensity modulated radiation therapy (IMRT) in our center from October 2007 to October 2013 were analyzed retrospectively. They were divided into 2 groups. Group NACT (76 patients) was treated with neoadjuvant chemotherapy followed with concurrent chemoradiation therapy, and group CCRT (71 patients) was treated with concurrent chemoradiation therapy. NACT received 2 cycles of neoadjuvant chemotherapy with the TP (docetaxel 75 mg/m2 d1 + cisplatin 80 mg/m2 d1) or TPF (docetaxel 75 mg/m2 d1 + cisplatin 80 mg/m2 d1 + 5-Fu 750 mg/m2d1-5) regimens, and then received 2 to 3 cycles of the platinum-based concurrent chemoradiation therapy. CCRT only received 2 to 3 cycles of concurrent chemoradiation therapy. Cumulative survival rate was analyzed by Kaplan Meier method. The log-rank test and Cox proportional hazard model was used for prognostic analyses. ResultsThe median follow-up time was 26 and 30 months in the NACT and CCRT groups, respectively. The 3-year LRRFS, DMFS, and OS for groups NACT and CCRT were 94.5%, 96.8%; 85.8%, 81.6% and 83.3%, 85.0%, respectively (P<.05). Age, sex, HGB, clinical stage, T stage, N stage, and radiation dose were included in prognostic analysis. For univariate analysis, only N stage was the adverse prognostic factor for 3-year DMFS (P<.05). However, for multivariate analyses, there was no statistical significance for these possible prognostic factors. The median follow-up time was 26 and 30 months in the NACT and CCRT groups, respectively. The 3-year LRRFS, DMFS, and OS for groups NACT and CCRT were 94.5%, 96.8%; 85.8%, 81.6% and 83.3%, 85.0%, respectively (P<.05). Age, sex, HGB, clinical stage, T stage, N stage, and radiation dose were included in prognostic analysis. For univariate analysis, only N stage was the adverse prognostic factor for 3-year DMFS (P<.05). However, for multivariate analyses, there was no statistical significance for these possible prognostic factors. ConclusionThere was no statistical significance for early survival results in NACT and CCRT groups; however, the neoadjuvant chemotherapy followed with concurrent chemoradiation therapy could increase the DMFS for locally advanced NPC patients to some degree. N stage was also a possible prognostic factor for the survival results of the local advanced NPC patients. There was no statistical significance for early survival results in NACT and CCRT groups; however, the neoadjuvant chemotherapy followed with concurrent chemoradiation therapy could increase the DMFS for locally advanced NPC patients to some degree. N stage was also a possible prognostic factor for the survival results of the local advanced NPC patients.

To explore the relationship between radiation-induced temporal lobe necrosis (RITLN) and the dose-volume for nasopharyngeal carcinoma (NPC) treated with intensity modulated radiation therapy (IMRT), the dose-volume outcome analysis based on a normal tissue complication probability (NTCP) model was observed. Six hundred ninety-five patients were retrospectively analyzed in the study. All of them were primarily diagnosed with NPC and treated with IMRT from January 2004 to January 2009 in 1 institution. Dose-volume of temporal lobe, treatment strategy, and clinical characteristics were systematically reviewed. RITLN was evaluated by two experienced radiologists on dynamic contrast enhancement magnetic resonance imaging separately. The relationship between RITLN and the dose of temporal lobe were analyzed, and dose-volume parameters that could predict RITLN were selected based on an NTCP model. A total of 59 of 695 patients were diagnosed with RITLN. The dose on necrosis region of temporal lobe was significantly higher than that in normal temporal lobe (P < .05). In addition, the temporal lobe necrosis regions were all located in a high-dose region in the primary plan. D2cc (the dose delivered to the 2 cubic centimeter volume, Gy) had the maximum area under curve (AUC) in above dose-volume parameters. Furthermore, our data suggested that it was an independent risk factor of RITLN according to multivariate analysis. Tolerance dose of temporal lobe (by which the RITLN incidence rate was less than 5% in 5 years), TD5/5 of D2cc was 61.97 Gy (the ROC [95% CI] = 60.74, 63.19), whereas TD50/5 of D2 cc was 73.26 Gy (95% CI = F 71.98, 74.46). Biological tolerance effective temporal lobe dose TD5/5 of D2cc was 60.31 Gy (95% CI = F 59.09, 61.54) and TD50/5 was 76.85 Gy (95% CI = F 75.74, 78.22). The incidence of radiation-induced temporal lobe necrosis was closely related to the dose of temporal lobe and corresponding volume. RITLN mainly due to higher radiation dose temporal lobe and results of the study suggested that the restriction dose (BED) of temporal lobe was D2cc <60.31 Gy.