Risk Groups For Overall Survival Research Articles

Relevance of different prognostic scores in primary CNS lymphoma in the era of intensified treatment regimens: A retrospective, multicenter analysis of 174 patients.

Treatment intensification (including consolidative high-dose chemotherapy with autologous stem cell transplantation [HDT-ASCT]) significantly improved outcome in primary central nervous system lymphoma (PCNSL) patients. We conducted a multicenter, retrospective analysis of newly diagnosed PCNSL patients, treated with intensified treatment regimens. The following scores were evaluated in terms of overall survival (OS) and progression-free survival (PFS): Memorial Sloan-Kettering Cancer Center (MSKCC), International Extranodal Lymphoma Study Group (IELSG), and three-factor (3F) prognostic score. Further, all scores were comparatively investigated for model quality and concordance. Altogether, 174 PCNSL patients were included. One hundred and five patients (60.3%) underwent HDT-ASCT. Two-year OS and 2-year PFS for the entire population were 73.3% and 48.5%, respectively. The MSKCC (p = .003) and 3F score (p < .001), but not the IELSG score (p = .06), had the discriminatory power to identify different risk groups for OS. In regard to concordance, the 3F score (C-index [0.71]) outperformed both the MSKCC (C-index [0.64]) and IELSG (C-index [0.53]) score. Moreover, the superiority of the 3F score was shown for PFS, successfully stratifying patients in three risk groups, which also resulted in the highest C-index (0.66). The comparative analysis of established PCNSL risk scores affirm the clinical utility of the 3F score stratifying the widest prognostic spectrum among PCNSL patients treated with intensified treatment approaches.

P13.06.A RADIOMIC FEATURES FOR RISK-STRATIFICATION IN PATIENTS WITH BRAIN METASTASES

Abstract BACKGROUND Recently, non-invasive characterization of brain tumors on MRI has emerged as a promising field of research. The identification of quantitative imaging biomarkers, also known as radiomics, may complement molecular characterization and thereby improve clinical management of neuro-oncologic patients. In this study we aimed to identify imaging predictors with improved performance over clinical parameters in order to stratify patients with brain metastases into high and low risk groups for overall survival (OS). MATERIAL AND METHODS 422 patients (allocated in a 3:1 ratio to a discovery [n=317] and test [n= 105] set) with first diagnosis of brain metastases from different primary tumors from two neuro-oncologic centers were included. In each patient, eight clinical features (age, gender, KPS, systemic disease status, presence of extracranial metastases, number of cerebral metastases, primary tumor and available individual prognostic molecular status eg HER2, BRAF,⋯) were gathered and a total of 321 radiomic MRI features (including shape, first-order and higher-order features) from cerebral MRI (contrast T1-weighted and apparent diffusion coefficient maps) were extracted. Radiomic and clinical features of patients in the discovery set were subjected to different machine learning models in order to classify patients into low- and high-risk groups for OS. By performing a comprehensive grid search the best machine learning model according to the macro-evaluated score and accuracy was identified and evaluated on the testing set. In addition, a subgroup analysis, based on the primary tumor entity was done and confusion matrices were calculated in order to evaluate final predictions. RESULTS With an extra trees classifier including all clinical and 30 radiomic features, we were able to stratify patients into a high- and low-risk group for OS (test set: macro-evaluated = 0.60, accuracy = 0.67). The best performing model was a gradient boosting model only including clinical features (test set: macro-evaluated = 0.62, accuracy = 0.72), while the radiomic features alone led to the poorest results (test set: macro-evaluated = 0.52, accuracy = 0.63). Interestingly, in the subgroup of melanoma patients, the predictive power of radiomic features outperformed the clinical and the combined (radiomics and clinical) features. With a prediction solely based on radiomic features, 80% and 67% of patients were correctly classified into the low- and the high-risk group for OS, respectively. CONCLUSION In conclusion, we found that in the entire study population of patients with brain metastases radiomic features did not allow for a better prediction of the clinical outcome compared to the clinical parameters alone. However, in the subgroup of melanoma patients the predictive power of radiomic features alone was superior compared to clinical features or all features combined.

A novel nomogram integrated with PDL1 and CEA to predict the prognosis of patients with gastric cancer.

This study aimed to develop a prognostic nomogram for patients with gastric cancer (GC) based on the levels of programmed death 1 ligand 1 (PDL1) and carcinoembryonic antigen (CEA). The nomogram was developed using data from a primary cohort of 247 patients who had been clinicopathologically diagnosed with GC, as well as a validation cohort of 63 patients. Furthermore, the nomogram divided the patients into three different risk groups for overall survival (OS)-the low-risk, middle-risk, and high-risk groups. Univariate and multivariate Cox hazard analyses were used to determine all of the factors included in the model. Decision curve analysis and receiver operating characteristic (ROC) curves were used to assess the accuracy of the nomogram. The Kaplan-Meier survival analysis revealed that metastasis stage, clinical stage, and CEA and PDL1 levels were predictors for progress-free survival (PFS) and OS of patients with GC. Metastasis stage, clinical stage, and CEA and PDL1 levels were found to be independent risk factors for the PFS and OS of patients with GC in a multivariate analysis, and the nomogram was based on these factors. The concordance index of the nomogram was 0.763 [95% confidence interval (CI) 0.740-0.787]. The area under the concentration-time curve of the nomogram model was 0.81 (95% CI 0.780-0.900). According to the decision curve analysis and ROC curves, the nomogram model had a higher overall net efficiency in forecasting OS than clinical stage, CEA and PDL1 levels. In conclusion, we proposed a novel nomogram that integrated PDL1 and CEA, and the proposed nomogram provided more accurate and useful prognostic predictions for patients with GC.

Predictive power of deep-learning segmentation based prognostication model in non-small cell lung cancer.

The study aims to create a model to predict survival outcomes for non-small cell lung cancer (NSCLC) after treatment with stereotactic body radiotherapy (SBRT) using deep-learning segmentation based prognostication (DESEP). The DESEP model was trained using imaging from 108 patients with NSCLC with various clinical stages and treatment histories. The model generated predictions based on unsupervised features learned by a deep-segmentation network from computed tomography imaging to categorize patients into high and low risk groups for overall survival (DESEP-predicted-OS), disease specific survival (DESEP-predicted-DSS), and local progression free survival (DESEP-predicted-LPFS). Serial assessments were also performed using auto-segmentation based volumetric RECISTv1.1 and computer-based unidimensional RECISTv1.1 patients was performed. There was a concordance between the DESEP-predicted-LPFS risk category and manually calculated RECISTv1.1 (φ=0.544, p=0.001). Neither the auto-segmentation based volumetric RECISTv1.1 nor the computer-based unidimensional RECISTv1.1 correlated with manual RECISTv1.1 (p=0.081 and p=0.144, respectively). While manual RECISTv1.1 correlated with LPFS (HR=6.97,3.51-13.85, c=0.70, p<0.001), it could not provide insight regarding DSS (p=0.942) or OS (p=0.662). In contrast, the DESEP-predicted methods were predictive of LPFS (HR=3.58, 1.66-7.18, c=0.60, p<0.001), OS (HR=6.31, 3.65-10.93, c=0.71, p<0.001) and DSS (HR=9.25, 4.50-19.02, c=0.69, p<0.001). The promising results of the DESEP model were reproduced for the independent, external datasets of Stanford University, classifying survival and 'dead' group in their Kaplan-Meyer curves (p = 0.019). Deep-learning segmentation based prognostication can predict LPFS as well as OS, and DSS after SBRT for NSCLC. It can be used in conjunction with current standard of care, manual RECISTv1.1 to provide additional insights regarding DSS and OS in NSCLC patients receiving SBRT. While current standard of care, manual RECISTv1.1 correlated with local progression free survival (LPFS) (HR=6.97,3.51-13.85, c=0.70, p<0.001), it could not provide insight regarding disease specific survival (DSS) (p=0.942) or overall survival (OS) (p=0.662). In contrast, the deep-learning segmentation based prognostication (DESEP)-predicted methods were predictive of LPFS (HR=3.58, 1.66-7.18, c=0.60, p<0.001), OS (HR=6.31, 3.65-10.93, c=0.71, p<0.001) and DSS (HR=9.25, 4.50-19.02, c=0.69, p<0.001). DESEP can be used in conjunction with current standard of care, manual RECISTv1.1 to provide additional insights regarding DSS and OS in NSCLC patients.

A prognostic score from a multicentric retrospective analysis of patients affected by sarcoma with metachronous lung metastases undergoing metastasectomy.

Despite the lack of evidence-based on prospective randomized studies, surgery has become the cornerstone of the treatment in patients with pulmonary oligometastatic sarcomas. Our study aimed to construct a composite prognostic score for metachronous oligometastatic sarcoma patients. A retrospective analysis was performed on data patients who underwent radical surgery for metachronous metastases in sixresearch institutes from January 2010 to December 2018. The log-hazard ratio (HR) obtained from the Cox model was used to derive weighting factors for a continuous prognostic index designed to identify differential outcome risks. A total of 251 patients were enrolled in the study. In the multivariate analysis, a longer disease-free interval (DFI) and a lower neutrophil-to-lymphocytes ratio (NLR) were predictive of a better overall survival (OS) and disease-free survival (DFS). A prognostic score was developed based on DFI and NLR data, identifying 2 risk class groups for DFS (3-years DFS 20.2% for the high-risk group [HRG]and 46.4% for the low-risk group [LRG] [<0.0001]) and 3 risk groups for OS (3 years OS 53.9% for the HRG vs. 76.9% for the intermediate-risk group and 100% of the LRG (p < 0.0001)). The proposed prognostic score effectively predicts outcomes for patients with lung metachronous oligo-metastases from the surgically treated sarcoma.

The comparison between albumin-bilirubin grade and Child-Pugh grade for assessing the prognosis of hepatocellular carcinoma after thermal ablation: a propensity score-matched analysis.

BackgroundThermal ablation is one of the first-line treatments for patients with hepatocellular carcinoma (HCC) with a massive potential of long-term cure. However, it is disadvantaged by the relatively high rate of postoperative recurrence. In this study, a comparison was performed in the prognostic performance between albumin-bilirubin (ALBI) grade and Child-Pugh (C-P) grade for those HCC patients undergoing thermal ablation.MethodsA total of 467 patients undergoing thermal ablation as the initial therapy were recruited and retrospectively analyzed. Then, an investigation was carried out into the prognostic values of overall survival (OS) and recurrence free survival (RFS) through ALBI grade and C-P score, separately. Besides, propensity-score matching (PSM) was performed to adjust the significantly different patient characteristics for ALBI grade based on C-P A patients.ResultsIn the multivariate analysis, both ALBI and C-P grades were treated as independent prognostic factors for OS. Cox multivariate analysis was conducted to reveal that the independent risk factors for postoperative HCC recurrence included ALBI grade 2-3, multiple tumors, tumor size ≥2 cm, and age ≥65 years. Following PSM (1:1), both ALBI-1 and ALBI-2 groups were found to be well matched in all of the parameters involved. Depending on the ALBI grade (113 ALBI-1 vs. 113 ALBI-2), C-P A patients were reclassified into two risk groups for OS (P=0.0058) and RFS (P=0.012), respectively.ConclusionsThe conventional C-P classification, despite its necessity, is less effective in assessing preoperative liver dysfunction for HCC patients and subjected to some limitations. Characterized by simplicity and objectivity, the ALBI grade has demonstrated its greater prognostic value than C-P grade for HCC patients who underwent thermal ablation.

The Predictive Power of a Deep-Learning Segmentation Based Prognostication (DESEP) Model in Non-Small Cell Lung Cancer

Previous work has reported that imaging features captured by a deep learning tumor segmentation algorithm correlates with overall and disease-specific survival, even without direct supervision using survival information. The current study aims to create a model to predict survival outcomes for non-small cell lung cancer (NSCLC) after treatment with stereotactic body radiotherapy (SBRT) using deep-learning segmentation-based prognostication (DESEP).The DESEP model was trained using imaging from 108 patients with NSCLC with various clinical stages and treatment histories. The model generated predictions based on unsupervised features learned by a deep-segmentation network from computed tomography imaging to categorize patients into high and low risk groups for overall survival (DESEP-predicted-OS), disease specific survival (DESEP-predicted-DSS), and local progression free survival (DESEP-predicted-LPFS). Serial assessments were also performed using auto-segmentation based volumetric RECIST and computer-based unidimensional RECIST methods.There was a high concordance between DESEP-predicted-LPFS risk category and manually calculated RECISTv1.1 (φ = 0.544, P = 0.001). Neither the auto-segmentation based volumetric RECIST nor the computer-based unidimensional RECIST correlated with RECISTv1.1 (P = 0.081 and P = 0.144, respectively). While RECISTv1.1 correlated with LPFS (HR = 6.97,3.51-13.85, c = 0.70, P < 0.001), it could not provide insight regarding DSS (P = 0.942) or OS (P = 0.662). In contrast, the DESEP-predicted methods were predictive of LPFS (HR = 3.58, 1.66-7.18, c = 0.60, P < 0.001), OS (HR = 6.31, 3.65-10.93, c = 0.71, P < 0.001) and DSS (HR = 9.25, 4.50-19.02, c = 0.69, P < 0.001).Deep-learning segmentation-based prognostication can predict LPFS as well as OS, and DSS after SBRT for NSCLC. It can be used in conjunction with RECISTv1.1 to provide additional insights regarding DSS and OS in NSCLC patients receiving SBRT.

Functional Assessment (FA) in Multiple Myeloma (MM) Patients Predicts Survival: Use of Most Valuable Comorbidity Scores and Functional Tests in a German Study Group (Deutsche Studiengruppe Multiples Myelom [DSMM]) Incentive

Introduction: Treatment options in Multiple Myeloma (MM) patients (pts) have substantially improved, but pts present with heterogeneous health status. Since the incidence of MM is increasing with age and the majority of pts is above 60 years, it is important to assess additional diseases and comorbidities. Measuring frailty, especially in older pts allows to identify those with an increased risk of treatment complications and impaired progression-free- (PFS) and overall-survival (OS). As MM interventions and therapy can be tailored, not only older, but also “intermediate-old” and younger MM pts may profit from such assessments. Therefore, a brief assessment of pts' functional, physiological and psycho-cognitive status was performed within this DSMM incentive, using 12 functional assessment (FA) tests and 5 established Comorbidity Indices (CI). The aim was to determine the most essential stratification for MM, to develop a meaningful, time-effective assessment arsenal for true estimation of pts' fitness and avoidance of both under- and overtreatment.Methods: We prospectively performed a FA in 301 consecutive newly diagnosed MM pts treated between 2011 and 2017. 12 functional tests comprised of the Karnofsky Performance Status (KPS), physician- and patient-rated fitness, Activity of daily living (ADL), Instrumental Activity of daily living (IADL), pain (VAS), Geriatric Depression Scale (GDS), Mini-Mental State Examination (MMSE), Timed Up and Go Test (TUGT), malnutrition, SF-12-Qualtiy of Life (QoL) Physical (PCS) and Mental Health Composite Scale (MCS). Additionally, 5 established CI were assessed: 1. International Myeloma working group (IMWG)-frailty index, 2. Revised Myeloma Comorbidity Index (R-MCI), 3. Charlson Comorbidity Index (CCI), 4. Hematopoietic Cell Transplantation-specific Comorbidity Index (HCT-CI) and 5. Kaplan-Feinstein-Index (KFI). In order to identify most valuable functional tests and CI, Kaplan Meier estimation and Cox regression for OS with backward variable selection were performed in order to determine which tests provided additional value for both IMWG-frailty index and R-MCI.Results: Pt characteristics were typical for tertiary centers. Median OS from the time point of assessment and median follow-up were 58 and 19 months, respectively. CI that grouped pts into significantly different risk groups for OS were the IMWG-frailty index, R-MCI and CCI, whereas both HCT-CI and KFI remained insignificant. Those CI that differentiated best between different risk groups were the IMWG-frailty index and R-MCI. 3-year-OS for good-, intermediate and high-risk pts using the R-MCI was 90%, 74% and 43% (p=0.0006) and via IMWG-frailty index 92%, 83% and 57% (p=0.0001), respectively. The 3-year-OS for good- vs. high-risk pts via CCI was 90% vs. 64% (p=0.0063). The 12 functional tests that significantly differentiated pts into different risk groups were the KPS with 3-year-OS for good- vs. high-risk pts of 89% vs 60% (p=<0.0001), physician-rated fitness of 80% vs. 66% (p=0.0085), patient-rated fitness of 82% vs. 64% (p<0.0036), ADL of 80% vs. 66% (p=0.0159), IADL of 78% vs. 48% (p=0.0036), pain of 78% vs. 66%, MMSE of 78% vs. 47% (p=0.0001), TUGT of 79% vs. 63% (p=0.0168) and SF-12 PCS of 86% vs. 66% (p=0.0091). GDS, malnutrition and SF-12 MCS tests proved insignificant. Backward selection with both IMWG-frailty index and R-MCI forced therein revealed 4 other tests to significantly complement these CI, namely the KPS, ADL, MMSE and SF-12 PCS.Conclusion: Our results suggest that our abbreviated FA, consisting of 2 CI (IMWG-frailty index and R-MCI) and 4 functional tests (KPS, ADL, MMSE, SF-12 PCS) constitutes a feasible, time-effective and highly diagnostic test set to estimate fitness and vulnerability in MM. Next steps include the automatic inclusion of the abbreviated assessment results into our MM-tumor board tool and pts' reports for high availability in medical discussions and therapy selection. Rather than focusing on pts' chronological age alone, our analysis aims to ultimately facilitate treatment decision and allow individualized treatment. DisclosuresLarocca:Janssen: Honoraria; Celgene: Honoraria; Bristol-Myers Squibb: Honoraria; Amgen: Honoraria. Engelhardt:German Cancer Aid (#11424): Other: Educational Grant; Janssen Cilag GmbH: Other: Educational Grant; Celgene GmbH: Other: Educational Grant; Amgen GmbH: Other: Educational Grant.

Evaluation of risk-stratification using gene expression assays in patients with breast cancer receiving neoadjuvant chemotherapy.

576 Background: Among patients with hormone receptor-positive, human epidermal growth factor receptor 2 (HER2)-negative breast cancer, several prospective studies investigated various gene expression assays, such as 21-gene recurrence score (21 RS) and 70-gene signature (70 GS), to identify a subgroup of patients with pathologic complete response (pCR) from neoadjuvant chemotherapy. However, in the absence of large prospective trials to validate such findings, the National Comprehensive Cancer Network guideline does not recommend the routine adoption of such assays in the setting of neoadjuvant therapies. To address this knowledge gap, we performed an observational cohort study to compare pCR and survival outcomes based on these assays. Methods: The National Cancer Database (NCDB) was queried for female patients diagnosed between 2010 and 2017 with stage I-III breast cancer who underwent neoadjuvant chemotherapy and either 70 GS or 21 RS. Logistic multivariable analysis (MVA) was performed to identify variables associated with pCR. Cox MVA was performed to evaluate overall survival (OS). Subgroup analyses were performed among patients with favorable hormone receptor status (hormone receptor-positive, HER2-negative) and with RS ≥26 instead of RS ≥31. Results: A total of 3,009 patients met our inclusion criteria, with 2,075 (n = 1,287 for RS &lt; 31, n = 788 for RS ≥31) and 934 (n = 175 for low risk, n = 759 for high risk) patients who underwent 21 RS and 70 GS, respectively. The median follow up was 48.0 months (interquartile range 32.2-66.7). On logistic MVA for all patients, those with a high risk from 70 GS or with RS ≥31 were more likely to have pCR. When compared to RS ≥31, a high risk from 70 GS was not associated with pCR. However, among those with favorable hormone receptor status, similar findings were noted, except that those with a high risk group from 70 GS were less likely to have pCR compared to those with RS ≥31. On Cox MVA for all patients, pCR was associated with improved OS. While RS ≥31 was associated with worse mortality, a high risk from 70 GS was not. No interaction was observed between pCR and risk groups for OS in both groups (interaction p = 0.23 for 70 GS, p = 0.66 for 21 RS). When analyses were repeated using a high risk group from 21 RS defined as RS ≥26, similar findings were noted, except that having favorable hormone receptor status and RS ≥26 was not associated with pCR when compared to the high risk from 70 GS. Conclusions: To our knowledge, this is the largest study using a nationwide oncology database suggesting that high recurrence risk groups in both assays were associated with pCR and that pCR was associated with improved survival. For those with favorable hormone receptor status, RS ≥31 may be a more selective prognostic marker. Further studies would be warranted to investigate the role of gene expression assays in the setting of neoadjuvant chemotherapy.

Geriatric nutritional risk index as an independent prognostic factor in locally advanced nasopharyngeal carcinoma treated using radical concurrent chemoradiotherapy: a retrospective cohort study.

Nutritional status is a key factor influencing the prognosis of patients with cancer. The Geriatric Nutritional Risk Index (GNRI) has been used to predict mortality risk and long-term outcomes. In this study, we aimed to evaluate the predictive value of pretreatment GNRI in patients with nasopharyngeal carcinoma (NPC). A total of 1,065 patients with biopsy-proven non-disseminated nasopharyngeal carcinoma were included. Based on a cutoff value of pretreatment GNRI, patients were divided into two groups (low ≤107.7 and high >107.7). Combining GNRI and baseline Epstein-Barr virus (EBV) DNA, all patients were further stratified into three risk groups, namely, high-risk (high EBV DNA and low GNRI), low-risk (low EBV DNA and high GNRI), and medium-risk (except the above) groups. Multivariate analyses were performed using the Cox proportional hazards model to assess the predictive value of the GNRI. Among the 1,065 patients, 527 (49.5%) and 538 (50.5%) were divided into low and high GNRI groups, respectively. Within a median follow-up of 83 months, patients with a high GNRI score exhibited significantly higher overall survival (OS), progression-free survival (PFS), and distant metastasis-free survival (DMFS) compared to those with low GNRI scores (P<0.05). Multivariate analyses revealed that high GNRI is an independent prognostic factor for OS and PFS (hazard ratio, HR, 0.471, 95% CI, 0.270-0.822, P=0.008; HR 0.638, 95% CI, 0.433-0.941, P=0.023, respectively). Using a combination of baseline GNRI and EBV DNA, a satisfying separation of survival curves between different risk groups for OS, PFS, DMFS was observed. The survival rates of patients in the high-risk group were significantly lower than those in the low- and medium-risk groups (all P<0.001). The combined classification was demonstrated to be an independent prognostic factor for OS and PFS after adjustment using multivariate analysis. Pretreatment GNRI is an independent prognostic factor for NPC patients. The combination of baseline GNRI score and EBV DNA level improved the prognostic stratification of NPC patients.

A prognostic nomogram based on LASSO Cox regression in patients with alpha-fetoprotein-negative hepatocellular carcinoma following non-surgical therapy

BackgroundAlpha-fetoprotein-negative hepatocellular carcinoma (AFP-NHCC) (< 8.78 ng/mL) have special clinicopathologic characteristics and prognosis. The aim of this study was to apply a new method to establish and validate a new model for predicting the prognosis of patients with AFP-NHCC.MethodsA total of 410 AFP-negative patients with clinical diagnosed with HCC following non-surgical therapy as a primary cohort; 148 patients with AFP-NHCC following non-surgical therapy as an independent validation cohort. In primary cohort, independent factors for overall survival (OS) by LASSO Cox regression were all contained into the nomogram1; by Forward Stepwise Cox regression were all contained into the nomogram2. Nomograms performance and discriminative power were assessed with concordance index (C-index) values, area under curve (AUC), Calibration curve and decision curve analyses (DCA). The results were validated in the validation cohort.ResultsThe C-index of nomogram1was 0.708 (95%CI: 0.673–0.743), which was superior to nomogram2 (0.706) and traditional modes (0.606–0.629). The AUC of nomogram1 was 0.736 (95%CI: 0.690–0.778). In the validation cohort, the nomogram1 still gave good discrimination (C-index: 0.752, 95%CI: 0.691–0.813; AUC: 0.784, 95%CI: 0.709–0.847). The calibration curve for probability of OS showed good homogeneity between prediction by nomogram1 and actual observation. DCA demonstrated that nomogram1 was clinically useful. Moreover, patients were divided into three distinct risk groups for OS by the nomogram1: low-risk group, middle-risk group and high-risk group, respectively.ConclusionsNovel nomogram based on LASSO Cox regression presents more accurate and useful prognostic prediction for patients with AFP-NHCC following non-surgical therapy. This model could help patients with AFP-NHCC following non-surgical therapy facilitate a personalized prognostic evaluation.

Importance of metastatic volume in prognostic models to predict survival in newly diagnosed metastatic prostate cancer.

To explore the prognostic importance of metastatic volume in a contemporary daily practice cohort of patients with newly diagnosed metastatic hormone-naive prostate cancer (mHNPC) and to develop a pragmatic prognostic model to predict survival for these patients. Since 2014, 113 patients with newly diagnosed mHNPC were prospectively registered. Statistical analysis was performed using SPSS 25.0™ with two-sided p value < 0.05 indicating statistical significance. Univariate and multivariate cox regression analyses were performed to identify prognostic risk factors. Kaplan-Meier method with log-rank statistics was constructed to analyze difference in survival in the prognostic groups. Model performance was assessed using the Concordance-index (C-index) and cross-validated in R v3.4.1. High-volume mHNPC (HVD) was defined as the presence of visceral metastasis or ≥ 4 bone metastases with ≥ 1 appendicular lesion. Multivariate analysis identified HVD (p = 0.047) and elevated alkaline phosphatase (ALP) (p = 0.018) as independent prognostic risk factors for overall survival (OS). Consequently, three prognostic groups were created: a good (no risk factors), intermediate (1 risk factor) and poor prognosis group (2 risk factors). Median OS for the good, intermediate and poor prognosis group was not reached, 73 and 20months (95%CI 9-31months with p < 0.001 and Correspondence-index of 0.78), respectively. We developed a pragmatic and qualitative prognostic model consisting of three prognostic risk groups for OS in a daily practice cohort of patients with newly diagnosed mHNPC. Independent prognostic risk factors included in the model were HVD and abnormal ALP.

Prediction of Overall Survival Based on Isocitrate Dehydrogenase 1 Mutation and 18F-FDG Uptake on PET/CT in Patients With Cerebral Gliomas.

This retrospective study aimed to correlate F-FDG uptake on PET/CT with isocitrate dehydrogenase enzyme isoform 1 (IDH1) mutation in patients with cerebral gliomas. Hierarchical interactions between factors affecting overall survival (OS) were also examined. In 59 patients with glioma, the ratio of the SUVmax of a glioma to the SUVmean of the contralateral cortex (G/C ratio) on F-FDG PET/CT and the presence of IDH1 mutation were correlated. The prognostic value of clinicopathologic factors and G/C ratio for OS were assessed using a Cox proportional hazards model and classification and regression tree models. The mean G/C ratio of IDH1-mutant tumors was significantly lower than that of IDH1 wild-type tumors (0.73 vs 1.14, P = 0.004). In multivariate analysis, IDH1-mutant and G/C ratio were significant for OS. The classification and regression tree modeling identified 3 risk groups for OS (group 1: IDH1 mutant [hazard ratio, 0.2]; group 2: G/C ratio ≤0.8 with IDH1 wild type [hazard ratio, 0.83]; group 3: G/C ratio >0.8 with IDH1 wild type [hazard ratio, 1.9]) (overall P < 0.001). The mean OS was 37.0 months in group 1, 28.6 months in group 2, and 20.7 months in group 3, respectively, showing significant differences among the groups (group 1 vs group 2: P = 0.023, group 2 vs group 3: P = 0.049, group 1 vs group3: P < 0.001). F-FDG uptake of IDH1-mutant gliomas was significantly lower than that of IDH1 wild-type gliomas. IDH1 mutation was the most important factor in identifying patients with the best prognosis, whereas increased F-FDG uptake provided additional prognostic information for predicting poor OS among patients with IDH1 wild-type gliomas.

Comparison of Prognostic Indices in Japanese Patients with Diffuse Large B-cell Lymphoma in the Yonago Area

Several prognostic indices for diffuse large B-cell lymphoma (DLBCL) have been developed. Which index is appropriate for Japanese patients with DLBCL treated in real-world practice is unknown. The prognostic performances of the original international prognostic index (IPI), age-adjusted IPI, National Comprehensive Cancer Network-IPI, elderly IPI and revised IPI were compared using patients with DLBCL treated in a single institute in the Yonago area in Japan. From 2005 through 2015, 182 patients with de novo DLBCL were treated with chemotherapy in Tottori University Hospital; 154 (85%) patients received R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine and prednisolone) although full dose was administered in 63 (35%) patients. The median age of the patients was 71 years (range 18 to 91). Three-year overall survival rate was 71.8% (95% CI, 64.1% to 78.2%). All indices significantly discriminate risk groups for overall survival of the patients (P < 0.001). Although no statistical difference of performance was found among these indices, the best scores of model fit/discrimination measures were beaten out by age-adjusted IPI, the simplest and three-factor model. Age-adjusted IPI is still usable in real-world practice while a better predictive model is desired for Japanese patients with DLBCL.

A Simple Score, Based on Geriatric Assessment, Predicts Survival in Elderly Newly Diagnosed Chronic Lymphocytic Leukemia Patients

BACKGROUNDClinical outcome of patients with chronic lymphocytic leukemia (CLL) is often poor due to the high prevalence of comorbidities as well as functional impairment that characterize this population. However, a consensus on the definition and measurement of frailty is lacking.The present analysis was designed with the aim to develop a simple and easily applicable frailty score to predict overall survival. PATIENTSOn a retrospective basis, a geriatric assessment (GA) has been performed in 113 CLL patients older than 65 years firstly diagnosed after January 2000 at our institution. GA included the Katz Activity of Daily Living (ADL), the Lawton Instrumental Activity of Daily Living (IADL) and the Cumulative Illness Rating Scale (CIRS).Median age of patients was 71 years (range,65-90) and 32.7% were older than 75 years. According to the Rai staging system 48 (42.4%) patients were in stage 0, 47 (41.6%) in stage I-II and 18 (16%)in stage III-IV. Eighteen patients (16%) had a CIRS score higher than 6, 17 (15%) an ADL score lower than 5 and 21 (18.5%) an IADL score lower than 6. RESULTSAfter a median follow-up time of 66 months (range,3-330) 29 (25,6%) out of 113 patients have died and 58 (51.3%) received therapy (i.e., 24 at the diagnosis and 34 after a median follow-up time of 29 months [range, 6-136 months]).Advanced age (HR=3.40), functional decline on ADL (HR=2.90) and IADL (HR=2.70), presence of comorbidities as assessed by CIRS score (HR=2.09), ECOG performance status (HR=4.94) and Rai clinical staging (stage 0 vs I-II, HR=3.65; stage 0 vs III-IV, HR=9.09) were all variables significant in univariate analysis. In a multivariate analysis which did not included Rai clinical staging a higher risk of death was observed for patients older than 75 years (HR=3.11; 95% CI, 1.60-6.06; P=0.001), with ADL lower than 5 (HR 3.50; 95% CI: 1.50-8.40; P=0.02) and CIRS higher than 4 (HR=2.09; 95% CI,1.08-4.05;P=0.03).An additive frailty score based on the integer part of HRs (i.e., 1 point for HR 1.1-1.9; 2 points for HR 2.0-2.9) was then calculated. By combining the risk scores (range, 0-8) for these variables, patients were stratified according to a cutoff finder analysis into 3 distinctive risk groups for overall survival : fit (score = 0, 53,3%), intermediate-fit (score=1-5; 41.5%) and frail (score= 6-8,10%). Median overall survival of patients belonging to fit, intermediate-fit and frail group was as follows: fit, 174 months [95% CI:126-222]; intermediate-fit, 104 months [95% CI:82-127]; frail, 37 months [95% CI:3-71] (P<0.0001)(Fig. 1). The predictive accuracy of our score using Harrell c-index was 0.70 (95% CI:0.53-0.87). Finally, in a head-to-head comparison with Rai clinical staging, by Cox multivariate analysis, our frailty score retained its prognostic significance (fit vs intermediate-fit, HR,3.41 [95% CI,1.63-7.15], P=0.001; fit vs frail, HR, 12.06 [95% CI,3.33-43.67], P<0.0001). CONCLUSIONSThis study represents the first attempt to develop an additive scoring system for CLL based on functional status, comorbidities and age. GA is a sensitive predictor of clinical outcome and could be used for selecting patients to treat with novel therapies. [Display omitted] DisclosuresMolica:Gilead Sciences: Speakers Bureau; Roche Italy: Membership on an entity's Board of Directors or advisory committees; Abbvie: Membership on an entity's Board of Directors or advisory committees; Jansen: Membership on an entity's Board of Directors or advisory committees.

Hematopoietic Stem Cell Transplantation with Reduced Intensity Conditioning in Older Patients with Myelodysplastic Syndromes

Allogeneic hematopoietic stem cell transplantation (HSCT) can be a curative therapeutic option for patients with myelodysplastic syndromes (MDSs). However, the role of allogeneic HSCT is still controversial for older MDS patients.Here we report the post-transplant outcomes in 88 older MDS patients (≥ 60 years) that were transplanted at MD Anderson Cancer Center with reduced intensity conditioning (RIC) from 2005 through 2015. In this cohort, the conditioning regimen were 1) intravenous busulfan in a dose calculated to target an average daily systemic exposure, represented by the area under the concentration versus time curve (AUC) of 4000 μMol-min ± 10% or 100 mg/m2 in combination with fludarabine 40 mg/m2 for 4 days or 2) melphalan 100-140 mg/m2 as a single dose with fludarabine 40 mg/m2 for 4 days. Donors were HLA matched related (MRD, n=48), matched unrelated (MUD, n=30) and haploidentical (n=10). Graft versus host disease (GVHD) prophylaxis in MRD and MUD recipients was tacrolimus (0.015 mg/kg) and methotrexate (5 mg/m2, 1,3,6,11 days). MUD recipients also received anti-thymocyte globulin (ATG) (4 mg/kg, in divided doses given on days -3, -2,-1) for GVHD prophylaxis. Post-transplant cyclophosphamide (50 mg/kg, 3-4 days), tacrolimus (0.015 mg/kg) and mycophenolate mofetil (MMF) (15 mg/kg) were used for GVHD prophylaxis in haploidentical donors.The median age was 65 (range, 60-77, Table 1). The study cohort had high-risk features including 30 patients (34%) with therapy-related MDS (tx-MDS) and 24 (27%) with monosomal karyotype-positive (MKpos) defined by Breems et al. Histological subtype at diagnosis was RAEB-1 and -2 in 33 (38%) and CMML in 14 (16%) patients. Hematopoietic stem cell comorbidity index (HCT-CI) was ≥3 in 45 (51%) patients. The median bone marrow blast count at HSCT was 3% (range, 0%-13%) in 87 patients.The median follow up of 36 survivors was 32 months (range, 6-98). Overall survival (OS), cumulative incidence of progression and transplant-related mortality (TRM) at 3 years was 41%, 26% and 35% respectively as presented in Table 2.Univariate analyses using Cox’s proportional hazards regression analysis showed that MKpos was the only prognostic factor for the cumulative incidence of progression. MKpos patients had higher incidence of progression compared with patient with good risk cytogenetics (including normal karyotype (n=23) and del(5q), del(12p), del(20q) (n=4)) (HR=9.5, 95% CI, 2.2-41, p=0.003) and with MKneg (HR= 3.3, 95% CI, 1.3-8.6, p=0.01). There was no significant difference between MKneg and CN (p=0.2). The univariate analyses showed that MKpos, MKneg, higher serum ferritin level at HSCT and HCT-CI≥3 were associated with inferior OS. Blast count at HSCT and age>65 were associated with worse survival but did not reach statistical significance.For multivariate analyses, we performed classification and regression tree (CART) analyses that enabled us to identify 3 risk groups for OS: 1) Poor risk group including patients with MKpos and BM blast >2% (n=12) with 3 year OS of 8%, 2) Good risk group including good risk cytogenetics with HCT-CI≤2 (n=14) with 3-year OS of 92%, and 3) intermediate risk group including rest of the group (n=55) with 3-year OS of 34% (Figure 1).In conclusion, our results show that allogeneic HSCT with RIC is a feasible option in older MDS patients with OS of 41% at 3-years. Results are best in patients with low HCT-CI and good risk cytogenetics. TRM and disease progression remain major causes of failure and innovative strategies with better disease control while not increasing TRM are urgently needed in this cohort. [Display omitted] [Display omitted] [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

The Significance of Serum Albumin Level and Platelet Count in Patients with Diffuse Large B-Cell Lymphoma in the Context of an Enhanced International Prognostic Index (NCCN-IPI)

Introduction The International Prognostic Index (IPI) is used to predict survival in patients with diffuse large B-cell lymphoma (DLBCL). The IPI includes five factors: age, stage, LDH, performance status, and extranodal sites. However, studies suggest that certain blood markers not included in the IPI, such as CRP, albumin, absolute lymphocyte count (ALC), and platelet count, predict survival. Recently, an enhanced IPI (NCCN-IPI) was proposed and showed excellent prognostic performance. It uses the same factors as the IPI, but refines the impact of age and LDH and the definition of extranodal disease. Few studies have assessed the prognostic impact of blood markers other than the five factors of the IPI in the context of the NCCN-IPI. Here, we analyzed the prognostic role of simple blood markers in the NCCN-IPI.Patients and Methods Data from consecutive DLBCL patients diagnosed between January 2004 and June 2014 were retrospectively analyzed. Patients receiving rituximab and anthracycline-based chemotherapy were included in the analysis. Values considered positive were CRP >1 mg/dl, albumin <3.5 g/dl, ALC <1,000/µl, or platelets <100,000/µl. The primary endpoint was overall survival (OS), assessed using the Kaplan-Meier method. The log-rank test and Cox regression analysis were used to assess the prognostic value of each variable.Results A total of 398 patients (median age, 69 years (range, 21-104)) were included. Of these, 124 (31.2%) had an ECOG performance status >1, 221 (55.5%) had advanced stage disease, and 223 (56.1%) had a high or high-intermediate NCCN-IPI score (>3). Pretreatment CRP, albumin, ALC, and platelet count values were positive in 215 (54%), 127 (31.9%), 144 (36.2%), and 31 (7.8%), respectively. The median follow-up was 34.8 months (range, 0.5-134) and the 5-year OS of the entire cohort was 68.8%. Univariate analysis identified CRP [hazard ratio (HR) 3.15; 95% confidence interval (CI) 2.07-4.80, p<0.001), albumin (HR 4.46; 95% CI 2.96-6.7, p<0.001), ALC (HR 2.40; 95% CI 1.62-3.54, p<0.001), platelet count (HR 4.69; 95% CI 2.87-7.67, p<0.001), and NCCN-IPI score >3 (HR 5.05; 95% CI 3.00-8.51, p<0.001) as significant for OS. NCCN-IPI, albumin, and platelet count remained significant in multivariate analysis (NCCN-IPI: HR 3.14; 95% CI 1.77-5.60, p<0.001; albumin: HR 2.41; 95% CI 1.36-4.27, p=0.002; platelets: HR 2.77; 95% CI 1.66-4.61, p<0.001), whereas CRP and ALC did not (CRP: HR 0.91; 95% CI 0.51-1.66, p=0.78; ALC: HR 1.33; 95% CI 0.87-2.03, p=0.18). Based on albumin and platelet count values, patients were divided into three risk groups for OS: group 1 included albumin- and platelet-negative patients (n=247); group 2 included either albumin- or platelet-positive patients (n=128); and group 3 included both albumin- and platelet-positive patients (n=23) (82.6%, 50.2% and 12.9%, respectively; p<0.001) (Figure 1A). The association of these risk groups with OS independently of the NCCN-IPI was evaluated. For patients with low or low-intermediate NCCN-IPI, the 5-year OS for group 1 (n=151), group 2 (n=22), and group 3 (n=2) was 89.2%, 76.8%, and 0%, respectively (p<0.001) (Figure 1B). For patients with high or high-intermediate NCCN-IPI, the 5-year OS for groups 1 (n=96), 2 (n=106), and 3 (n=21) was 72.6%, 44.3%, and 14.6%, respectively (p<0.001) (Figure 1C). Assessment of the prognostic impact of albumin and platelet count in the elderly led to the exclusion of the low risk category (score 0-1) among those older than 60 years (n=291) because of their age score points (>1 points) in the NCCN-IPI; thus, the ability of NCCN-IPI to differentiate risk groups might be diminished in the elderly. For low-intermediate (n=97), high-intermediate (n=118), and high (n=76) NCCN-IPI patients, the 5-year OS for the elderly according to the NCCN-IPI was 80.3%, 64.8% and 42.5%, respectively (p<0.001) (Figure 2A). The combination of albumin and platelet count also predicted survival in the elderly: the 5-year OS for groups 1 (n=169), 2 (n=103), and 3 (n=19) was 77.7%, 50.1%, and 0%, respectively (p< 0.001) (Figure 2B).Conclusion Despite the improved performance of the NCCN-IPI, albumin and platelet count were independent prognostic factors and predicted survival even in the elderly. Considering albumin and platelets may improve the accuracy of recently established prognostic models. Prospective studies should evaluate the prognostic impact of albumin and platelets. [Display omitted] [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Center for International Blood and Marrow Transplant Research Chronic Graft-versus-Host Disease Risk Score Predicts Mortality in an Independent Validation Cohort

We previously reported a risk score that predicted mortality in patients with chronic graft-versus-host disease (CGVHD) after hematopoietic stem cell transplantation (HCT) between 1995 and 2004 and reported to the Center for International Blood and Marrow Transplant Research (CIBMTR). We sought to validate this risk score in an independent CIBMTR cohort of 1128 patients with CGVHD who underwent transplantation between 2005 and 2007 using the same inclusion criteria and risk score calculations. According to the sum of the overall risk score (range, 1 to 12), patients were assigned to 4 risk groups (RGs): RG1 (0 to 2), RG2 (3 to 6), RG3 (7 to 8), and RG4 (9 to 10). RG3 and RG4 were combined, as RG4 accounted for only 1% of the total cohort. Cumulative incidences of nonrelapse mortality (NRM) and probability of overall survival were significantly different between each RG (all P < .01). NRM and overall survival at 5 years after CGVHD for each RG were 17% and 72% in RG1, 26% and 53% in RG2, and 44% and 25% in RG3, respectively (all P < .01). Our study validates the prognostic value of the CIBMTR CGVHD RGs for overall survival and NRM in a contemporary transplantation population. The CIBMTR CGVHD RGs can be used to predict major outcomes, tailor treatment planning, and enroll patients in clinical trials.

Comprehensive analysis of genetic alterations and their prognostic impacts in adult acute myeloid leukemia patients

To clarify the cooperative roles of recurrently identified mutations and to establish a more precise risk classification system in acute myeloid leukemia (AML), we comprehensively analyzed mutations in 51 genes, as well as cytogenetics and 11 chimeric transcripts, in 197 adult patients with de novo AML who were registered in the Japan Adult Leukemia Study Group AML201 study. We identified a total of 505 mutations in 44 genes, while only five genes, FLT3, NPM1, CEBPA, DNMT3A and KIT, were mutated in more than 10% of the patients. Although several cooperative and exclusive mutation patterns were observed, the accumulated mutation number was higher in cytogenetically normal AML and lower in AML with RUNX1-RUNX1T1 and CBFB-MYH11, indicating a strong potential of these translocations for the initiation of AML. Furthermore, we evaluated the prognostic impacts of each sole mutation and the combinations of mutations and/or cytogenetics, and demonstrated that AML patients could be clearly stratified into five risk groups for overall survival by including the mutation status of DNMT3A, MLL-PTD and TP53 genes in the risk classification system of the European LeukemiaNet. These results indicate that the prognosis of AML could be stratified by the major mutation status in combination with cytogenetics.

MD Anderson Scoring System (MDACC) Predicts Outcomes After Hematopoietic Stem Cell Transplantation (HSCT) Better Than Other Prognostic Classifications In MDS

MDS is a heterogeneous group of hematopoietic stem cell disorders. Various prognostic models have been established to categorize patients with MDS including the International Prognostic Scoring System (IPSS), the Revised-IPSS (r-IPSS) and MDACC Scoring System. In this analysis, we compared those three classification schemas for their outcome predictability after HSCT.We analyzed 291 MDS patients with a median age of 55 (interquartile range (IQR) 47-60.7 years) who underwent HSCT between January 2001 and December 2011. Histology by WHO classification included RA/RARS 48 (16.5%), RCMD 28 (9.6%), RAEB-1 59 (20.2%), RAEB-2 63 (21.7%), MDS unclassified 67 (23%), and CMML 26 (9%). Of 291, 117 patients (40.2%) had therapy related MDS (t-MDS). Conditioning regimen was myeloablative in 201 patients (69.1%) and reduced intensity in 90 patients (30.9%). Donors were matched related (MRD), matched unrelated (MUD), mismatched (MMD) in 131 (45%), 114 (39.2%) and 46 (15.8%) patients respectively. Risk categorization was performed by IPSS, r-IPSS and MDACC scoring systems at the time of diagnosis. IPSS, r-IPSS and MDACC scoring systems could be assessed in 239 (82.1%), 241 (82.8%) and 231 (79.4%) patients respectively.The median follow up time of 109 survivors was 45 months. The median time from diagnosis to HSCT was 7.3 months (IQR 4.6-12.4 months). Three-year overall survival (OS) was 38.1% (95%CI 32.3-43.9) with 3-year event free survival (EFS) of 34.2% (95%CI 28.4-40). Cumulative relapse incidence (RI) at 3-year was 28.8% (95%CI 23.3-34.5). Cumulative incidence of treatment related mortality (TRM) at 3 year post-transplant was 27.9% (95%CI 22.6-33.6).In univariate analysis, IPSS and r-IPSS were able to differentiate 2 risk groups for OS and EFS. High risk group per IPSS and very high risk group per r-IPSS had lower OS with hazard ratio (HR) of 2.4 to 3.1, lower EFS with HR of 2.2 to 2.7. While IPSS could not predict RI, very high risk group by r-IPSS had higher RI with HR of 3.6 compared with lower risk groups. Both IPSS and r-IPSS did not identify different risk groups for TRM.On the other hand, MDACC scoring system was able to identify 4 different risk groups for EFS and OS in univariate analysis. Three-year OS was 68%, 46.1%, 30.3% and 11.4% for patients with MDACC risk score of 0-4, 5-6, 7-8 and ≥9 respectively (p<0.001) (figure1). Three-year EFS with MDACC risk score of 0-4, 5-6, 7-8 and ≥9 was 61.7%, 40.8%, 28.1% and 7.4% respectively (p<0.001). For RI and TRM, only MDACC risk scores of ≥9 was associated with poor outcomes with 3-year RI of 38.9% and 3-year TRM of 41.7% compared with 13.3% and 15.5% in risk scores of 0-4 (p=0.01 and p=0.01 respectively).In multivariate analysis, MDACC score, matched unrelated and mismatched donors were associated with inferior OS (table1).As a summary, MDACC risk scoring system for MDS better differentiates prognostic groups than IPSS or r-IPSS. Considering the high frequency of t-MDS among transplanted MDS patients, we propose that MDACC scoring system should be used for prognostic classification for hematopoietic transplantation. Disclosures:No relevant conflicts of interest to declare.