Lee, Ju‐Whei

12102 Background: Allostatic load (AL) conceptualizes the effects of chronic psychosocial adversity on physiologic dysregulation. To date, studies have shown an association between elevated AL and higher disease-specific and overall mortality among cancer patients; however, none have focused on multiple myeloma (MM) patients. We aim to understand the relationship between baseline AL, symptom burden, and mortality among patients enrolled in the E1A11 therapeutic trial in MM. Methods: ECOG-ACRIN E1A11 was a phase III RCT comparing induction with Bortezomib (Arm A) versus Carfilzomib (Arm B) in conjunction with Lenalidomide +Dexamethasone. AL included 7 biomarkers: BMI, alkaline phosphatase, creatinine, C-reactive protein, white blood cell count, albumin and creatinine clearance. AL7 was a composite summary score with a point was assigned for each biomarker value in the highest quartile, except for albumin and creatinine clearance, where a point was assigned for values in the lowest quartile. Endpoints included symptom burden at baseline and ̃1 month, non-completion of induction therapy, and overall survival (OS). Functional Assessment of Cancer Therapy Multiple Myeloma (FACT-MM) items assessed patient-reported symptom burden, including fatigue (item HI7), pain (GP4), and bother by side effects of treatment (GP5) on a 5-point Likert scale. Multivariable logistic regressions assessed the effect of AL7 (ranging 0-7) on high-pain, -fatigue, and -bother (QOL score > = 3 vs < 3), and non-completion of induction therapy. The effect of AL7 on OS was assessed using multivariable Cox regression. Regression covariates included study arm, age, sex, race, ECOG performance status, and the target symptom burden score at baseline. Results: The study cohort included 1087 patients. Mean baseline AL7 was 1.8 (±1.4). In adjusted analysis, a unit increase in AL7 was associated with a greater odds of high pain (OR 1.15, 95%CI [1.04-1.27]) and high fatigue (OR 1.19, 95%CI [1.07-1.32]) at baseline, which did not persist at ̃1 month (pain OR 0.96, 95%CI [0.84-1.10]; fatigue OR 1.03, 95%CI [0.91-1.16]). There was no association between AL7 and high side effect bother at baseline (OR 1.06, 95% CI [0.83-1.35]) or at ̃1 month (OR 1.06, 95%CI [0.90-1.24]). There was no association between AL7 and induction non-completion (OR 1.07, 95%CI [0.96-1.18]). Notably, each unit increase in AL7 was associated with higher mortality (HR 1.26, 95%CI [1.14-1.39]). Conclusions: Despite its association with fatigue and pain at baseline, AL7 was not associated with these symptoms at ̃1 month nor induction non-completion. However, elevated baseline AL7 was associated with poorer OS. AL composite score at baseline, which we interpret as a measure of physiological dysregulation associated with adverse social factors, may have implications on clinical outcomes within clinical trials despite presumed equal treatment access.

e19136 Background: Patient reported side effect bother is needed to understand tolerability of cancer drugs. We examined whether increase in side effect bother from baseline to cycle 7 as measured by a single item (GP5) from the FACT-G, “I am bothered by side effects of treatment”, was associated with early treatment discontinuation (ETD) in a multiple myeloma (MM) trial. Methods: Data were drawn from the induction phase of ECOG ACRIN E1A06, a two arm, phase 3 trial conducted in patients with untreated MM. The induction phase occurred over twelve, 28-day cycles. GP5 was assessed at registration and cycle 7. ETD was defined as completion of less than 12 cycles of treatment. We tested whether increase in side effect bother on the GP5 from baseline to cycle 7 was associated with months to ETD in two ways. At each timepoint, GP5 was rated with these response options: “Not at all”, “A little bit”, “Somewhat”, “Quite a bit”, “Very Much”. First, we stratified Kaplan Meier curves by patients with versus without substantially increased side effect burden from registration to cycle 7 on the GP5, defined as an increase of > 2 response categories; Cox regression was used to calculate a hazard ratio. Second, we fit a joint model of the GP5 change trajectory on months to ETD that substituted the estimated slope of a longitudinal logistic regression model into the hazard function of a Cox model. In this model, GP5 was dichotomized as 0 = “Not at all/”A little bit”; 1 = “Somewhat”/ “Quite a bit”/ “Very Much”. Results: 159 patients were followed during induction for a median of 11 months (range: 0.2-16.0). 10 patients (6%) reported a substantial increase on GP5 from registration to cycle 7. After 16 months, these patients had a significantly higher hazard of ETD as estimated by Kaplan Meier methods: 40% experienced ETD vs. 24% of patients without substantial GP5 increase [hazard ratio (HR): 3.08 (95% CI: 1.18-8.02)]. In the joint model, the effect of GP5 on hazard of ETD was larger [HR: 9.56 (95% CI: 2.41-37.82)]. Conclusions: This study found initial evidence that increase in side effect bother as measured by the GP5 predicts ETD and may therefore reflect treatment intolerability in cancer drug trials. The effects found in this study will be tested for replication in additional ECOG-ACRIN trials with diverse designs, treatments, and cancer types. Clinical trial information: NCT00602641 .

e19132 Background: The Functional Assessment of Cancer Therapy–General has an item about patient tolerability of treatment: “I am bothered by side effects of treatment” (GP5). We examined the predictive value of this single item for early treatment discontinuation in clinical trials. Methods: GP5 level prior to protocol therapy (rated using a 5-point Likert scale) and treatment start/end dates and off treatment reason data at each treatment phase were drawn from five phase III clinical trials conducted by ECOG-ACRIN. In the present analysis, GP5 was dichotomized as 0 = “Not at all”/“A little bit” and 1 = “Somewhat”/“Quite a bit”/“Very Much”. Early treatment discontinuation was defined either as receiving less than protocol specified cycles of treatment when maximum cycles specified in the protocol (E1A06 induction, E1912 induction, E1609 induction, E1105 induction, E5103 adjuvant), analyzed using logistic regression via odds ratio [OR]), or treatment cessation for reasons other than progressive disease or death when treatment continued until progression or intolerability (E1A06 maintenance, E1912 maintenance, E1609 maintenance, E1105 maintenance), analyzed using Cox proportional hazard model via hazard ratio [HR]. Results: GP5 prior to treatment was significantly associated with early discontinuation of E1A06 maintenance, E1609 maintenance, E1912 maintenance, and E1912 induction. No significant association was found for other therapies examined in the study. Conclusions: High GP5 level prior to treatment is associated with higher likelihood of early treatment discontinuation in patients who have received previous treatment. The limited predictive value of GP5 for treatment naïve patients is more limited, serial on-treatment assessment should be considered in this setting. Clinical trial information: NCT00602641 . [Table: see text]

6016 Background: A germline mutation in let-7 complementary site 6 (LCS6) within the KRAS 3'-untranslated region (rs61764370, the KRAS-variant: TG/GG) is known to associate with poor outcome and drug resistance in various cancers compared to the wild type allele (TT). We examine the prognostic significance of the KRAS-variant in recurrent/metastatic (R/M) head and neck squamous cell carcinoma (HNSCC). Methods: The KRAS-variant was determined in 116 tumor DNA samples from HNSCC patients enrolled in 3 clinical trials and a tissue collection study using a previously validated PCR-based assay. Results: KRAS-variant status could be determined in 108/116 (93%) samples and an allele frequency of TG/GG was 28.7%. These results were correlated with patient demographics, p16/human papillomavirus (HPV) status and clinical outcome. There was no association between p16/HPV status and the KRAS-variant status (Fisher’s exact test, p=1.0). The KRAS-variant was associated with poor progression-free survival in patients treated with cisplatin+/-cetuximab (log-rank p=0.002) but this association was not observed in docetaxel/bortezomib treated patients (log-rank p=0.89). Conclusions: KRAS-variant is a potentially promising biomarker of poor prognosis and a predictive biomarker of cisplatin resistance in R/M HNSCC. Prospective validation is warranted. Clinical trial information: NCT00003809.

6028 Background: Abnormalities in EGFR signaling targets are associated with C resistance but no biomarker of C resistance has been identified in HNSCC. We hypothesized that cases with loss of PTEN protein expression (PTEN null) or PIK3CA mutation would display C resistance in HNSCC. Methods: E5397 was a phase III trial of CDDP plus P or CDDP plus C and enrolled 117 eligible and evaluable pts. PIK3CA and PTEN were analyzed for 52 and 67 consented pts, respectively. PTEN expression (PTEN Cell Signaling Technology, Cat. 9559) was determined by automated quantitative analysis (AQUA) on the PM-2000 (HistoRx, New Haven) using a cutpoint generated in 5 HNSCC tissue microarrays, each consisting of HNSCC as well as positive (small intestine, median AQUA score 2833.2) and negative controls (breast and colon carcinoma, median AQUA score 205.5). A cutpoint of 570 provides 100% specificity, 100% sensitivity, and identified 30% of the HNSCCs as PTEN null, consonant with the literature. The 3 most common PIK3CA mutations (E542K and E545K in exon 9 and H1047R in exon 20) were determined by BEAMing (Inostics, Heidelberg, Germany). Response, overall survival (OS) and progression-free survival (PFS) were compared between PTEN null or PIK3CA mutated pts and all others. Log rank and multivariable Cox proportional hazards modeling were used to calculate p values. Results: 23/67 (34%) tumors were PTEN null and 2/52 (4%) had PIK3CA mutations (E542K and E545K). Both tumors with PIK3CA mutation had PTEN expression. No statistically significant differences in response, OS or PFS were noted in this small sample. However, among PTEN expressing/PIK3CA WT pts, median PFS increased to 4.2 months (m) for C (N=22) from 2.9 m for P (N=26) (Wald p=0.07), compared with 4.6 m for C (N=12) and 3.5 m for P (n=13) among the PTEN null/PIK3CA mutated (Wald p=0.60). Conclusions: The PTEN loss or PIK3CA mutation signature warrants further investigation as a predictor of C resistance.

Abstract Background: Novel therapies are required to improve the outcome of patients with AML. New agents are asked to demonstrate an overall survival (OS) benefit before qualifying for FDA approval. The long duration of clinical trials required in order to achieve this endpoint hampers quick evaluation of candidate therapies, including novel agents. Identification of reliable surrogate endpoints for OS in AML is needed. Here we compare the results of therapy for patients with untreated AML ages 16-60 years on the Eastern Cooperative Oncology Group 1900 trial (E1900) of induction chemotherapy followed by consolidation and autologous transplant in order to evaluate the validity of an event free survival (EFS) endpoint as a surrogate for OS. Methods:OS was measured from randomization for induction therapy to death from any cause (censored at last contact). EFS was measured from randomization to induction treatment failure, relapse after compete response (CR), or death in remission (censored at last contact). Hazard ratios (HR) were computed using Cox proportional hazards models. The association between EFS and OS was evaluated using the Kendall tau-a rank correlation for censored data. Results:There were657 patients enrolled of which 426 patients relapsed or had induction treatment failure before death or date of last contact. Median EFS and OS were 8.0 months (95% CI, 6.3 to 9.7 months) and 23.6 months (95% CI, 16.9 to 23.6 months), respectively. With a median follow-up of 80.1 months, there is a statistically significant correlation between EFS and OS (Kendall tau-a = 0.467, 95% confidence interval (CI) = (0.425, 0.510), p<0.001). This correlation was similarly significant at a median follow-up of 25.2 months (Kendall tau-a = 0.361, 95% CI (0.323, 0.400), p <0.001) when the E1900 trial was originally reported (Fernandez et al. NEJM 2009). Key findings reported based on the original OS endpoint are similar when analyzed with an EFS endpoint (Table 1). High-dose daunorubicin (90 mg/m2) (DNR 90) confers both an EFS and OS benefit in patients aged < 50 years and patients with intermediate cytogenetic risk, and does not confer an EFS or OS benefit in older patients and patients with unfavorable cytogenetic risk, on univariate analysis. Divergent results are only seen in the small subset of favorable cytogenetic risk patients, where DNR 90 conferred an OS benefit (p=0.027) without an EFS benefit (p=0.32). Both EFS and OS endpoints consistently reflect the impact of mutation status on survival. The presence of a FLT3-ITD or DNMT3A mutation has a negative impact on both EFS and OS while an IDH2 mutation has a favorable impact on EFS and OS. The presence of a NPM1 mutation confers a favorable impact on EFS and OS in patients who received DNR 90 and did not impact EFS or OS in patients receiving standard-dose daunorubicin (45 mg/m2) (DNR 45). The presence of an IDH1 mutation does not impact EFS or OS. Conclusions:The results of E1900 demonstrating superiority of DNR 90 in AML induction in patients up to age 60 are concordant when using an EFS or OS endpoint. This is true for the group as a whole as well as for subgroups for which targeted agents are in development (FLT3/IDH2 inhibitors). Further investigation of whether EFS is a reliable surrogate for OS is warranted in AML. If confirmed, its use as a primary endpoint could be adopted by regulatory agencies in order to allow more rapid completion of clinical trials in AML and bring new therapies to AML patients in a timely fashion. Table 1. Results of E1900 based on an EFS endpoint versus an OS endpoint. Subgroup N OS HR (DNR 90/DNR 45) & 95% CI Wald P EFS HR (DNR 90/DNR 45) & 95% CI Wald P DNR 45 DNR 90 Age < 50 yrs ³ 50 yrs 188 142 172 155 0.66 (0.50, 0.85) 0.81 (0.62, 1.06) 0.002 0.118 0.64 (0.50, 0.82) 0.86 (0.67, 1.10) 0.0004 0.23 Cytogenetic Favorable Intermediate Unfavorable 38 141 59 51 127 63 0.51 (0.28, 0.93) 0.68 (0.50, 0.92) 0.79 (0.54, 1.16) 0.027 0.012 0.225 0.76 (0.44, 1.31) 0.63 (0.47, 0.83) 0.72 (0.49, 1.05) 0.32 0.001 0.09 Subgroup N OS HR (MUT/WT) & 95% CI Wald P EFS HR (MUT/WT) & 95% CI Wald P FLT3-ITD WT MUT 456 147 1.62 (1.31, 2.01) <.0001 1.48 (1.21, 1.82) 0.0002 DNMT3A WT MUT 371 119 1.30 (1.03, 1.65) 0.03 1.23 (0.98, 1.54) 0.07 IDH1 WT MUT 465 36 0.88 (0.59, 1.33) 0.55 0.91 (0.62, 1.34) 0.64 IDH2 WT MUT 451 50 0.63 (0.43, 0.93) 0.02 0.68 (0.48, 0.97) 0.03 NPM1 DNR 45 DNR 90 245 257 0.84 (0.61, 1.16) 0.60 (0.41, 0.89) 0.30 0.01 0.90 (0.66, 1.22) 0.59 (0.41, 0.84) 0.49 0.004 Disclosures No relevant conflicts of interest to declare.

Abstract Background: Eastern Cooperative Oncology Group trial E1900 (E1900) showed that induction therapy with a high daily dose of daunorubicin (90 mg/m2) (DNR 90) improves survival in younger patients (pts) (<50 yrs) and intermediate (int) cytogenetic risk AML, but at 2 years of follow-up no benefit was seen in older pts (50-60 yrs), or in those with unfavorable cytogenetic risk or FLT3-ITD mutant AML (Fernandez et al. N Engl J Med 2009). Here we update results of E1900 with longer follow-up, focusing on the benefit of DNR 90 in cytogenetic and common molecular subgroups. Methods: Overall survival (OS) was measured from randomization for induction therapy to death from any cause (censored at last contact). Hazard ratios (HR) for death were computed using univariate and multivariable Cox proportional hazards models; multivariable Cox models were adjusted for sex, age, hemoglobin level, leukocyte count, platelet count, and cytogenetic profile. All conclusions regarding the impact of DNR 90, unless noted, are similar based on univariate and multivariable analysis. Results: Overall, 657 pts were enrolled with a median follow-up of 80.1 months. The HR for death in the DNR 90 group as compared with the standard-dose daunorubicin (45 mg/m2) (DNR 45) group was 0.74 (p=0.001). Pts <50 yrs benefited from DNR 90 (p = 0.002) while those >=50 yrs were not proven to benefit (p = 0.12). Pts with favorable (fav) and int. cytogenetic risk benefited from DNR 90 (p = 0.03 and p = 0.02, respectively). A benefit for pts with unfavorable cytogenetic risk was seen on multivariable analysis (p = 0.04). Impact of DNR 90 by mutation status: The 3 most common mutations were FLT3-ITD (24%), NPM1 (26%), and DNMT3A (24%). AML pts with any of these 3 mutations benefited from DNR 90 (p = 0.009, p = 0.002, and p = 0.02, respectively). FLT3-ITD pts who received DNR 90 had a 4-yr OS of 31%. Benefit was seen in pts age 50-60 with FLT3-ITD or NPM1 mutation (p = 0.02 and p = 0.04, respectively). No benefit of DNR 90 was seen in a small cohort of pts with MLL-PTD (p = 0.06). Benefit of DNR 90 in FLT3-ITD, NPM1, and DNMT3A mutant AML was confirmed in the int. cytogenetic risk group. Impact of DNR 90 on prognostic impact of NPM1: The presence of an NPM1 mutation conferred an improvement in OS in the DNR 90 group which was not seen in the DNR 45 group (p = 0.01 vs p = 0.3). This finding was confirmed in the int. cytogenetic risk group. Conclusion: With median follow-up of over 6 years on E1900, we confirm that DNR 90 improves outcome in pts with fav/int cytogenetic risk, DNMT3A or NPM1 mutant AML, or age < 50 (Patel et al. N Engl J Med 2012). Additionally, we now demonstrate that DNR 90 additionally benefits pts with FLT3-ITD AML, and pts with unfavorable cytogenetic risk, regardless of age. Moreover, we show that the favorable prognostic impact of the NPM1 mutation is only present when pts receive DNR 90. Given the benefit of DNR 90 across all cytogenetic risk groups as well as common molecularly defined subgroups of AML, DNR 90 should be the standard for all pts up to age 60 who are candidates for induction chemotherapy. Table HR for death by AML cohort. Subgroup N Univariate Model DNR 45 DNR 90 HR (DNR 90/DNR 45) & 95% CI Wald P All patients (n=657) Overall 330 327 0.74 (0.61, 0.89) 0.001 Age < 50 yrs³ 50 yrs 188 142 172 155 0.66 (0.50, 0.85) 0.81 (0.62, 1.06) 0.002 0.12 Cytogenetic Favorable Intermediate Unfavorable 38 232 59 51 212 63 0.51 (0.28, 0.93) 0.76 (0.61, 0.96) 0.79 (0.54, 1.16) 0.03 0.02 0.22 FLT3-ITD WT MUT 215 83 241 64 0.74 (0.59, 0.92) 0.61 (0.42, 0.89) 0.008 0.009 MLL-PTD WT MUT 290 16 296 15 0.70 (0.58, 0.86) 0.46 (0.21, 1.04) 0.0004 0.06 NPM1* WT MUT 180 65 192 65 0.70 (0.55, 0.89) 0.50 (0.32, 0.78) 0.003 0.002 DNMT3A WT MUT 177 61 194 58 0.66 (0.52, 0.85) 0.62 (0.41, 0.94) 0.001 0.02 * Statistically significant test of interaction (p<0.2) Figure 1A: Overall Survival by NPM1 Mutation Status and Treatment Arm Figure 1A:. Overall Survival by NPM1 Mutation Status and Treatment Arm Figure 1B: Overall Survival of FLT-ITD Mutant AML Patients by Treatment Arm Figure 1B:. Overall Survival of FLT-ITD Mutant AML Patients by Treatment Arm Disclosures No relevant conflicts of interest to declare.

Abstract Background Acute promyelocytic leukemia (APL) is commonly complicated by complex coagulopathy. The early use of all-trans retinoic acid (ATRA) appears to have reduced induction mortality; however, with modern regimens the risk of early hemorrhagic death remains at about 5-10% in clinical trials and up to 30% in population-based studies. Previously identified as markers of bleeding risk in this setting are: white blood cell count (WBC), peripheral blood blast count, platelet count, fibrinogen, prothrombin time (PT), activated partial thromboplastin time (aPTT) and creatinine. Nonetheless, uncertainty remains as to which ones are independent predictors. Materials and Methods The database for study was from APL patients enrolled on three large clinical trials that used ATRA for induction: APML3 (single arm of ATRA + idarubicin ± prednisone), APML4 (single arm of ATRA + idarubicin + arsenic trioxide + prednisone) and ECOG 2491 (intergroup I0129, consisting of daunorubicin + cytarabine vs ATRA). Known determinants of bleeding were assessed at baseline and considered as potential predictors of hemorrhagic death in the first 30 days of induction. In univariate analysis, using Wilcoxon Rank Sum and Fisher’s exact test, we looked at the association of the parameters mentioned above with hemorrhagic death within 30 days of induction. Results A total of 619 patients were enrolled: ECOG 2491 (intergroup I0129) - 394, APML3 - 101, and APML4 -124. Among these, thirty (4.8%) deaths from bleeding during the first 30 days (early deaths, ED) following induction were recorded. Baseline characteristics of the patients are shown in Table 1 and Table 2. Initial coagulation parameters (PT, aPTT and fibrinogen) were not available for the ECOG trial. Parameters with significant differences between the ED cases and the remaining of the cohort included WBC (median: 7.4 vs 2.4 /Kmcl respectively, p<0.001), peripheral blood blast count (median: 1.0 vs 0.2 /Kmcl respectively, p=0.049) and platelet count (median: 20 vs 33 /Kmcl respectively, p=0.002). Conclusions In this large cohort of approximately 600 APL patients enrolled on 3 large clinical trials, higher WBC, peripheral blast count and lower platelet count emerged as univariate predictors of early hemorrhagic death. Surprisingly, age, PT, aPTT, fibrinogen and creatinine clearance did not differ between the patients who died from bleeding and the remaining of the cohort. Multivariate analysis will clarify the relative strength of the significant predictors in assessing ED, and will be available, along with coagulation parameters for ECOG, at the time of presentation. These data hopefully will allow us to build a clinical prediction rule to identify patients at the highest risk of early hemorrhagic death. Such a stratification tool could prove useful in future trials testing novel approaches to treat the coagulopathy associated with APL. : Abstract 948. Table 1 Baseline Characteristics of Patients, Continuous Variables Variable Total Death from Bleeding within 30 days No Yes N Median (IQR) N Median (IQR) N Median (IQR) p-value Age (years) 619 40.0 (26.0 - 54.0) 589 40.0 (26.0 - 54.0) 30 47.0 (26.0 - 63.0) 0.15 White Blood Cell Count (K/mcL) 611 2.5 (1.2 - 8.2) 582 2.4 (1.1 - 7.6) 29 7.4 (3.1 - 20.7) <0.001 Peripheral blood blast count (K/mcL) 581 0.2 (0.0 - 2.2) 552 0.2 (0.0 - 1.9) 29 1.0 (0.1 - 5.7) 0.049 Hemoglobin (g/dL) 615 9.3 (8.0 - 10.5) 585 9.3 (7.9 - 10.5) 30 9.4 (8.2 - 10.7) 0.75 Platelets (K/mcL) 615 31.0 (17.0 - 56.0) 585 33.0 (17.0 - 57.0) 30 20.0 (13.0 - 34.0) 0.002 Prothrombin time (s) 184 16.0 (13.3 - 18.2) 179 16.0 (13.3 - 18.1) 5 16.7 (16.5 - 144.3) 0.15 Activated partial thromboplastin time (s) 225 29.0 (26.0 - 32.0) 216 29.0 (26.0 - 32.0) 9 31.2 (29.0 - 38.0) 0.07 Fibrinogen (mg/dL) 217 1.8 (1.3 - 2.6) 208 1.8 (1.3 - 2.6) 9 2.3 (2.0 - 2.5) 0.36 Creatinine Clearance (mg/dL)* 225 389 (261.2 - 511.6) 216 392 (264.9 - 512.4) 9 262 (219.2 - 432.0) 0.15 *calculated using the Jelliffe formula : Table 2 Baseline Characteristics of Patients, Categorical Variables. Variable Total Death from Bleeding within 30 days No Yes N % N % N % p-value ECOG Performance Status 579 93.54 552 93.72 27 90.00 0.50 Good(0-2) Poor(3-4) 36 5.82 33 5.60 3 10.00 Missing 4 0.65 4 0.68 0 0 FAB Classification 248 40.06 239 40.58 9 30.00 >0.95 M3 M3v 67 10.82 65 11.04 2 6.67 Missing 304 49.11 285 48.39 19 63.33 Disclosures No relevant conflicts of interest to declare.

Abstract Background Extramedullary disease (EMD) of acute myeloid leukemia (AML) was described decades ago but the incidence of this phenomenon and its prognostic impact are not clear. It is also uncertain whether every site of EMD has the same significance. This study explored these questions using a large cohort of AML patients treated on consecutive Eastern Cooperative Oncology Group (ECOG) frontline clinical trials. Methods For AML patients enrolled into ECOG clinical trials, the presence of EMD was captured at baseline on case report forms. From patients with newly diagnosed AML, age 15 and above, who were treated on 11 consecutive different clinical trials, we identified those with or without EMD, defined by physical examination, laboratory findings and imaging, without necessarily a biopsy. We used descriptive statistics to summarize patient and disease characteristics. Univariate analyses of potential prognostic factors were done. The Kaplan-Meier method was used to estimate median overall survival (OS) within each prognostic category and differences were explored using the log-rank test. Cox proportional hazards models were used to examine the effect of one-unit increases in continuous variables on OS and for multivariable analyses. Multivariate models were built using backward selection. Factors (or groups of factors) significant at the 0.10 level in univariate analyses were tested for inclusion in the model, and retained if they were significant at the 0.05 level. Results Of the 3,522 patients enrolled in 11 different AML clinical trials, we excluded 281; for diagnosis of other types of leukemia (n=220), no EMD evaluation at baseline (n=41) or no survival data (n=20). The overall incidence of EMD was 23.8% (770 out of the remaining 3,241 patients). The sites involved were: lymph nodes 367 (11.3%) patients, spleen 234 (7.2%), liver 173 (5.3%), skin 146 (4.5%), gingiva 104 (3.2%), central nerve system (CNS) 32 (1%), peripheral nerve system (PNS) 8 (0.2%) and other sites 134 (4.1%). In 65 cases (8.4%) EMD was confirmed by biopsy. Most of the patients (64.4%) had only one site of EMD, 163 (21.2%) had 2 sites, 75 (9.7%) - 3 sites, 28 (3.6%) - 4 sites, and 4 patients (0.5%) each had 5 or 6 sites. EMD patients compared to those without EMD; tend to be younger (median age 45.7 vs 52.9 years); male (57.7% vs 52%); poorer ECOG performance status (76.4% with ECOG 0-1 vs 85.9%) and with higher white blood cell count (WBC) at diagnosis (median of 41.6/µl vs 10.2/µl). In univariate analysis, having EMD was associated with a shorter OS (P=0.006). Examination of individual EMD sites revealed that skin (P=0.002), spleen (P=0.0002) and liver (P=0.0007) but not CNS (P=0.35), PNS (P=0.53), nodal involvement (P=0.85) and gingival hypertrophy (P=0.14), were associated with poorer OS. Using proportional hazards models for continuous factors, each additional site of EMD conferred a 9.4% increase in the risk of death. In a multivariable model, after adjusting for known prognostic factors (such as: age, WBC count and cytogenetic risk group), the presence of EMD, number of EMD sites and any specific EMD site were not independently prognostic. Of 165 patients with known favorable cytogenetics EMD was present in only 22 patients (13%). This group appeared to have a worse prognosis, but the numbers are too small for a definitive assessment (figure 1). Conclusions Extramedullary disease is common in newly diagnosed AML with an incidence of almost 24%, but CNS involvement is very rare (0.95%). In contrast to published data, no site of EMD was found to have an independent prognostic impact in multivariate analysis. This large dataset emphasizes the importance of evaluating large number of patients and considering all known risk factors in the multivariate analysis. Disclosures: No relevant conflicts of interest to declare.