ALBERT

Description: Abstract 4439 Background: Several reports have recently been published documenting the experience of patients with Philadelphia chromosome positive chronic myeloid leukemia in the chronic phase (CML-CP) who were switched from Glivec (IM), the β-crystalline form of imatinib mesylate, to non-identical alternate copies of imatinib mesylate (IMac) including those composed of the α-crystalline form. As a result of pre-clinical tests performed during the early development of IM, the β-crystalline form of imatinib mesylate was selected for further development due to its superior stability. In one case report evaluation, 126 Iraqi patients with CML-CP were switched from IM to an IMac that was the α-crystalline form of imatinib mesylate. At the time of switch to IMac, patients had previously received 400mg IM once-daily for an average of 50 months and were at least in complete hematologic response (CHR). Post-switch, patients were initially placed on the same IMac dose and followed-up on a monthly basis; more frequently in cases of poor response or adverse events. By 3 months post-switch, 22 (17%) patients had lost hematologic response and of these patients 18 (14%) and 4 (3%) had progressed to accelerated phase (AP) and blast crisis (BC), respectively. By 6 months post-switch, an additional 20 patients (16%; 33% total) had lost hematologic response. Safety assessment for IMac in this cohort has not been previously published. Objective: To assess the clinical safety and to estimate the survival of Iraqi CML patients switched from IM to IMac. Methods: Patient case records were retrospectively reviewed for safety findings as assessed by the treating physician. A previously published Markov transition-state model was used to compare projected life-years (LYs), progression-free life-years (PFLY), and quality-adjusted life-years (QALYs) of IM patients switched to IMac vs. patients not switched. Patients entered the model after a mean 50 months of IM therapy. At that time, based on the IRIS trial results, patients were considered to have CHR (4.7%), partial (6.5%) or complete (88.9%) cytogenetic response. Patients remaining on IM transitioned within these 3 responses levels and no hematologic response, AP, BC, and death according to the original model probabilities. For patients switched to IMac, transition rates were based exclusively on response rates observed in the Iraqi study (Scenario 1) or on the Iraqi study for the first 6 months, and thereafter based on the transition rates originally in the Markov model (Scenario 2). Utilities were from the original model. Results: Non-hematologic adverse event (AE) rates observed in patients who were switched to IMac are reported (Table). Patients remaining on IM were predicted to experience 15.7 LYs, 14.5 PFLYs, and 13.4 QALYs. Corresponding numbers for patients switched to IMac were 2.4 LYs, 1.1 PFLYs, 1.4 QALYs (in Scenario 1) and 11.4 LYs, 10.2 PFLYs, and 9.6 QALYs (in Scenario 2). Results were also sensitive to response distribution at model entry. Conclusions: The clinical safety, efficacy, and tolerability of IM have been established from more than 10 years of clinical experience. Recent introduction of generic versions of imatinib (including α-crystalline form) allows assessment of whether patients switching from IM to a generic version are able to achieve the same clinical outcomes as with the branded drug. The Iraqi case report is the largest cohort currently available for such review. This case report suggests that switching from IM to an IMac that does not have the safety and/or efficacy may result in substantial loss of LYs, PFLYs, and QALYs. Robust pharmacovigilance programs may need to be established to differentiate safety findings for originator drugs vs. substandard generic versions. Disclosures: Botteman: Novartis: Consultancy. Magestro:Novartis: Employment, Equity Ownership. Manley:Novartis Pharma AG: Employment. Zernovak:Novartis Pharmaceuticals Corp: Employment, Equity Ownership. Woodman:Novaris Pharmaceuticals Corp: Employment, Equity Ownership.

Description: Abstract 2765 In CML, achievement of major molecular response (MMR) is a significant prognostic factor as it has been shown to be associated with longer duration of complete cytogenetic response (CCyR) and long-term progression-free survival. In IRIS study, patients who achieved both CCyR and MMR showed higher progression-free survival rates, compared to those who had CCyR without MMR. Higher doses of imatinib are expected to yield higher CCyR and MMR rates, compared to standard dose of imatinib, and second-generation tyrosine kinase inhibitor, nilotinib also produces high CCyR and MMR rates in patients with CP CML who are resistant to imatinib. In this prospective study, the efficacy of nilotinib and high-dose imatinib was investigated in suboptimal molecular responders who received standard-dose imatinib as first-line therapy. Early CP CML patients who have achieved CCyR but no MMR after at least 18 months and up to 24 months (≥ 18 to ≤ 24 months) on first-line imatinib therapy at a daily dose of 400 mg were enrolled in this clinical trial, and informed consents were obtained from all patients prior to participation. In nilotinib arm, patients received oral dose of 400 mg BID (800 mg/day), and patients received 800 mg/day administrated as 400 mg BID in imatinib dose-escalation arm. To assess the drug efficacy, cytogenetics and RQ-PCR analysis were performed at regular intervals, and baseline mutational analysis was conducted for every patient with subsequent mutational analyses performed in patients demonstrating either lack of response or disease progression. Primary endpoint is to evaluate the cumulative MMR rates by 12 months, and secondary endpoints are to evaluate the cumulative CMR rates and time to and duration of MMR and CMR during further 24 month follow-up. Progression-free survival and safety profiles will also be assessed as secondary endpoints. Patients showing lack of response (lack of complete hematologic response (CHR) at 6 months, increasing WBC, no major cytogenetic response (MCyR) at 24 months), loss of response (loss of CHR or MCyR) or severe intolerance to treatment were allowed to crossover to the alternative treatment arm. With a data cut-off date of 18 Jul 2011, a total of 30 patients were randomized into nilotinib arm (n =13) or imatinib arm (n = 17), and 6 patients have crossed-over to nilotinib arm due to lack of response. With a median follow-up of 11 months (range, 0.2–28 mos), all patients have maintained CCyR without progression to advanced disease, and progressive decrease in BCR-ABL transcript levels was observed in all patients. Cumulative MMR rates at 20 months were significantly higher in nilotinib arm compared to imatinib dose-escalation arm (59.00% vs. 27.40%, P = 0.047), and patients treated with nilotinib also showed faster molecular response rates, with 5 patients achieving MMR within 3 months of nilotinib therapy. At the last follow-up, 7/13 (53.85%) and 2/11 (18.18%) patients achieved MMR in nilotinib arm and in high-dose imatinib arm, respectively, with 1 patient in nilotinib arm achieving 4-log reduction of BCR-ABL transcripts. Although toxicity was observed more frequently in imatinib dose-escalation arm, all patients currently maintain the initial dose (except 1 patient who interrupted imatinib therapy due to neurosurgical operation), and based on the toxicity data, no additional or serious adverse events were developed except for pre-existing toxicities before randomization. These preliminary results demonstrate that early intervention using nilotinib or dose escalation of imatinib could be recommended in suboptimal molecular responders, with nilotinib being more preferable. Through further clinical investigation on a large patient population and longer period of observation, efficacy and safety of early intervention of suboptimal molecular response using nilotinib or dose escalation of imatinib will be assessed. Updated data with longer follow-up duration will be presented in the meeting. Disclosures: Woodman: Novartis: Employment, Equity Ownership. Szczudlo:Novartis: Employment, Equity Ownership. Kim:Novartis: Employment.

Description: Background: Imatinib, nilotinib, and dasatinib are BCR-ABL tyrosine kinase inhibitors (TKIs), with different selectivity profiles, approved for the treatment of patients (pts) with Philadelphia positive chronic myeloid leukemia (Ph+ CML). Given the differences in their kinase selectivity profiles, the safety profiles of these agents also differ, particularly with regard to episodes of fluid retention. The incidence of pleural/pericardial effusions in resistant or intolerant CML pts who failed imatinib, or both imatinib and dasatinib, therapy was evaluated. Methods: The occurrence of pleural/pericardial effusions in resistant or intolerant CML pts following therapy with imatinib, or imatinib and dasatinib sequentially, was evaluated in 915 pts with Ph+ CML in chronic phase (CML-CP, 60%), accelerated phase (CML-AP, 23%), and blast crisis (CML-BC 17%) who entered the nilotinib compassionate use program between June 2006 and April 2008. At the time of medical review for compassionate use approval, safety information including the presence of or the history of pleural/pericardial effusions was collected along with dosing information for imatinib and dasatinib. Nearly all pts analyzed (94%) had not received nilotinib therapy prior to inclusion in the compassionate use program. Resistance and intolerance as well as CML phase were defined using similar criteria as previously reported in the nilotinib pivotal phase I/II study. Results: The median age was 52 years (range: 11 – 87 years); 22 pts (2%) were 18 years old or younger. Most pts (734; 80%) had received prior imatinib only (64% discontinued due to resistance, 20% with resistance and intolerance and 16% due to intolerance) and 170 pts received both imatinib and dasatinib (29% resistant; 17% with resistance and intolerance, 54% intolerant), with most dasatinib failures being due to toxicity. Information on past treatment is currently unavailable for 11 pts. Of pts who were pretreated with imatinib alone, less than 2% of pts (n = 11) had pleural and/or pericardial effusions reported. Among those pts pretreated with both imatinib and dasatinib, 51 pts (30%) had pleural and/or pericardial effusions, with 50 (98%) having pleural effusions, 10 of which were bilateral. Fifty-five percent of pts with pleural and/or pericardial effusions were in chronic phase. Of the pts with dasatinib-associated pleural/pericardial effusions; 11% were noted on daily doses 〉 140 mg, 29% on doses of 140 mg, 20% on doses of 100 mg, and 10% on doses of 〈 100 mg. There were 14 pts who developed pleural and/or pericardial effusions in which the dasatinib dose was not available. Of the 33 pts with pleural and/or pericardial effusions on daily 140 mg dasatinib, 16 pts (48%) discontinued dasatinib due to pleural effusion and 6 pts exhibited persistent pleural effusions despite dasatinib dose reductions. One pediatric patient (age 11 years) had a pleural effusion with once daily 80 mg dasatinib. Pleural/pericardial effusions occurred in all age groups: 〉50 years old – 6 pts, 51–60 years old – 10 pts; 61–70 years old – 13 pts; 71–80 years old – 16 pts and 〉81 years old – 4 pts (2 of unknown age). No pts who developed dasatinib-associated pleural/pericardial effusions had a history of pleural/pericardial effusions during imatinib therapy. Conclusions: This large dataset supports earlier reports that pleural/pericardial effusions are frequently reported among CML pts treated with dasatinib compared with imatinib. Furthermore, the occurrence of dasatinib-associated pleural/pericardial effusions may occur at any dose, even below the standard 140 mg/day dose and in some cases with doses less than 100 mg/day.

Description: Background: Moderate-to-severe anemia has been shown to adversely affect health outcomes in patients with cardiovascular disease (CVD). Less is known about the potential conjoint impact of mild anemia and CVD on frailty status in community-dwelling older adults. The objectives of this cross-sectional study were two-fold: (1) to assess the risk of frailty associated with mildly low Hb in community-dwelling older women with and without CVD, and (2) to test whether the association between Hb and frailty differed according to CVD status. Methods: Data from women 70–80 years old who participated in the baseline assessments of the Women’s Health and Aging Studies (WHAS) I and II were analyzed. Analysis was restricted to those with Hb 〉/= 10g/dL, and known frailty and CVD status. CVD was defined as angina, myocardial infarction, and/or congestive heart failure. Frailty was defined according to previously validated criteria; i.e., 3 or more of the following: slowness (walking speed10%, weakness (grip strength

Description: Background: AZA monotherapy has demonstrated improvement in OS in HR MDS, clinically meaningful and durable responses continue to be limited to a subset of patients (Silverman 2006). One obvious strategy is to develop doublet therapy with drugs that are effective monotherapy in HR MDS and can be administered effectively and safely in combination with AZA. Several agents have been combined with AZA in first line therapy for pts with HR-MDS achieving ORR and CR/PR rates that are comparable to AZA monotherapy (ORR 38% and CR/PR of 24%) (Ades ASH 2018; Sekeres JCO 2017). RAS and other signaling molecules in the Ras pathway are frequently mutated in HR MDS and are proposed to drive leukemic transformation (Takahashi 2013). Given that rigosertib interferes with the RAS-binding domain of RAF kinases and inhibits the RAS-RAF-MEK & the PI3Ks pathways (Athuluri-Divakar, Cell 2016), it is an attractive candidate for combination with AZA. Furthermore, in vitro combo of rigosertib with AZA synergistically inhibits growth and induces apoptosis of leukemic cells in a sequence-dependent fashion (Skiddan AACR 2006). We report updated results from a Phase II study in a subset of patients receiving oral rigosertib in combination with standard dose AZA as first line therapy for HR MDS. Methods: In the Phase II study 09-08, a total of 39 treatment-naïve HR MDS/RAEB-t patients received oral rigosertib in combination with standard dose AZA. Rigosertib was given at 840mg/day (560 mg in the morning & 280mg in the afternoon); or 1120mg/day (560 mg in the morning & 560mg in the afternoon or 840 mg in the morning and 280 mg in the afternoon) (Maniar ASH 2018). Responses were determined by 2006 IWG criteria including transfusion independence (56 days without PBC or PLT transfusions). Oral rigosertib was administered on D1-D21 of a 28D cycle. Parenteral AZA 75mg/m2/day was administered for 7days from D8. Patients were evaluable for response if they received 3 cycles of therapy. Results: Median age of the pts was 64 years (42-90). IPSS-R score at study entry was 9 were Intermediate, 8 were High and 17 were Very High Risk (VHR). In total 20 pts were transfusion-dependent at study entry. CR/PR responses were observed across all IPSS-R cytogenetic prognostic subgroups: Very Poor 60%, Poor 25%, Intermediate 37.5% & Good 25%. CR/PR responses were also observed across all higher IPSS-R prognostic risk categories: Very High 42%, High 17% and Intermediate 25%. A summary of clinical benefit/risk is provided in Table 1 below. Median duration of response was 12.2 mos (0.1-24.2+) and median duration of treatment was 8 mos (1.3-27.3). Time to first and best responses was 1/4 mos. Four pts continue to respond to therapy. 10 AEs led to D/C: urinary tract pain (2), and 1 each for urinary retention, hematuria, hydronephrosis, osteolysis, cerebral haemorrhage, WBC count decreased, neutrophil count decreased, & abd pain. The most frequent AE in table 1, are similar to AEs reported for both rigosertib and AZA as monotherapies, & the GU toxicities were mitigated using specific management guidelines. The majority of MDS pts who experienced AEs ≥Grade 2 were successfully managed and continued to receive the doublet on study. Conclusion: The efficacy (90% ORR & 34% CR) and safety of oral rigosertib and AZA in combination is favorable as first line therapy in pts with HMA naïve HR MDS and are comparable to historical results for standard dose AZA monotherapy (ORR 38% and

Description: Introduction: Anemia among the elderly population is often under-diagnosed or goes undetected. The incidence of anemia among this population is believed to be 4–6 times greater than reported, with an underlying condition found in over half of anemics, especially men and older individuals (J Am Geriatr Soc1997;45(7):825–31). Given the number of comorbidities in elderly patients a better understanding of the relationship between these underlying conditions and anemia is warranted. Methods: A retrospective analysis of a large health claims database (1997–2002) examined the prevalence of anemia and associated comorbidities in patients 〉/=65 years with 〉/=3 years of follow-up. Chronic anemia was defined using the WHO criteria (hemoglobin [Hb] 30 days and /=2 ICD-9 diagnosis codes or 〉/=1 CPT code for each condition, and categorized a priori as ‘linked’ [potential association with anemia (rheumatoid arthritis and/or lupus, CKD, CHF, untreated cancer and TB)] or ‘non-linked’ [not associated with anemia (diabetes, asthma and/or COPD, osteoarthritis, osteoporosis, hypertension and CVD]. Patients were categorized by number of comorbidities (0–1, 2–3, 4–5, 〉5) and age (65–74, 75–84, 〉/= 85). Results: The mean age among the study sample (chronic anemics: n=2729; non-anemics: n=7648) was 72.9 years (SD=5.9); 56.8% were women and 31.2% were African American. Mean Hb was similar for all groups (65–74 [n=1446]: 11.7±0.89; 75–84 [1080]: 11.7±0.94; 〉/=85[203]: 11.6±0.96, and 〉/=65[2729]: 11.7±0.91). Only 19.6% of anemics had an anemia diagnosis code. Patients with chronic anemia had significantly more comorbid conditions than those without anemia (p5 comorbidities, respectively). The same trend was observed across all age groups. Regardless of anemia status, the number of comorbid conditions increased with age among patients with and without anemia, however, patients with anemia were 2.5 times as likely to have 〉3 comorbidities compared to patients without anemia across all ages (OR=2.51, 95%CI=2.30–2.75). The number of linked comorbidities was observed to increase with age. Approximately 7% of non-anemic patients had 〉/=2 linked diseases, while almost 23% of chronic anemics had 〉/=2 linked comorbidities (p3 conditions: 19.5% vs. 11.0%, p

Description: Abstract 3826 Background: Imatinib is the current standard of care for chronic myelogenous leukemia (CML). Nilotinib is a highly potent and the most selective inhibitor of BCR-ABL. A Phase III multi-center, open label, randomized study (ENESTnd) was conducted comparing these two therapies in adult patients with newly diagnosed Philadelphia Chromosome positive (Ph+) CML in chronic phase. The primary endpoint analysis at 12 months demonstrated that major molecular response (MMR) was significantly improved with nilotinib 300 mg BID (44%) and nilotinib 400 mg BID (43%) compared to imatinib 400 mg QD (22%; p 〈 0.001). The discontinuation rate due to adverse events was lowest among the nilotinib 300mg BID treatment arm (5%) compared to 7% in the imatinib arm, and nilotinib 400 mg BID (9%). Based on the results of this clinical trial, nilotinib 300 mg BID was approved for initial use for CML-CP in the US. Aim: To evaluate the occurrence and rate of hospitalizations and time away from usual activities in this phase III trial. Methods: A total of 846 patients were randomized to receive nilotinib 300 mg BID (n=282), nilotinib 400 mg BID (n=281) or imatinib 400mg QD (n=283). Hospitalizations, defined as any visit to the hospital requiring an overnight stay, excluding pre-planned or elective surgery, were assessed throughout the study period. Overdispersed Poisson regression models were used to compare the days hospitalized per 1,000 patient-days on study. Patients were asked to report time-off, defined as average number of hours per week taken away from all usual activities due to CML and side effects of CML treatment over the past 4 weeks, at Baseline and at the end of Months 3 and 12. The Wilcoxon rank-sum test was used to compare the time off from usual activities at each assessment; and t-tests were used to evaluate the within-group changes in time off. Results: There were a total of 57 hospitalizations in the imatinib arm versus 48 hospitalizations in the nilotinib 300 mg BID arm, and 74 hospitalizations in the nilotinib 400 mg BID arm (Table). Descriptive statistics for length of stay (LOS) are presented in the Table. The hospitalization rate, expressed as hospital days per 1,000 patient days, was 47% higher in the imatinib arm compared to the nilotinib 300 mg BID arm (p=0.057) and 8% higher compared to the nilotinib 400 mg BID arm (p=0.68). Patients in the nilotinib 300mg BID arm had fewer stays and shorter LOS than the imatinib arm, whereas patients in the nilotinib 400mg BID arm had more stays than the imatinib arm but shorter LOS on average resulting in fewer total hospital days. The majority of hospitalizations (56%) in all three arms occurred within the first 9 months. Time off from usual activities, which began at an average level of 8–10 hours per week at Baseline, decreased in each arm, but the decrease did not significantly differ between arms (Table). Similar results were observed when patients reporting zero hours of time off were excluded from the analysis. There was no association between time off and age. Summary/conclusions: In patients with newly diagnosed CML-CP, nilotinib resulted in less hospital time compared to imatinib, although this difference did not reach statistical significance. Additionally, patients in all three treatment groups reported significant improvements from baseline in time off from usual activities. Disclosure: Beaumont: Novartis: Research Funding. Coombs:Novartis: Employment, Equity Ownership. Bollu:Novartis: Employment, Equity Ownership. Woodman:Novartis Oncology: Employment. Cella:Novartis: Research Funding.

Description: Abstract 2283 Introduction: A switch to the 2nd generation tyrosine kinase inhibitor nilotinib has been proven to be effective in case of resistance or intolerance to imatinib for the majority of patients with Ph+ chronic myeloid leukemia (CML) in chronic phase (CP). Besides mutations in the BCR-ABL kinase domain and various BCR-ABL-independent mechanisms, e.g. clonal evolution and activation of pathways bypassing BCR-ABL, efficacy of imatinib and nilotinib are dependent on intracellular drug levels, which are influenced by the activity of the efflux transporter protein multidrug resistance 1 (MDR1). Cell culture data suggest overexpression of MDR1 as a cause of resistance to nilotinib (Mahon et al., Cancer Res 2008). Moreover, Dulucq et al. (Blood 2008) report the association of MDR1 single nucleotide polymorphisms (SNPs) with the chance to achieve major molecular response (MMR) on first-line imatinib therapy. Aim: In order to allow risk stratification, we sought to elucidate molecular markers, e.g. MDR1 gene expression, BCR-ABL transcript burden, BCR-ABL mutation status and common SNPs in the MDR1 gene regarding their potency to predict therapy-related endpoints, such as MMR, complete cytogenetic response (CCyR), and progression-free survival (PFS) during second line therapy with nilotinib in imatinib-resistant CML-CP patients. Patients and Methods: A cohort of 83 imatinib-resistant patients in chronic phase CML treated with nilotinib was investigated within the AMN107A2101 phase I/II trial. Baseline BCR-ABL mutations were detected by D-HPLC and direct sequencing. MDR1 and BCR-ABL mRNA expression levels were determined by qRT-PCR using LightCycler™ technology, normalized against beta-glucuronidase (GUS) expression. BCR-ABL levels were standardized according to the international scale (IS). MDR1 SNPs (1236C〉T, 2677G〉T) were investigated using conventional sequencing. Log-rank tests were performed to compare the time to MMR (BCR-ABL IS ≤0.1%), CCyR, and PFS. Results: By 12 or 24 months, patients with MDR1/GUS ratios ≥2 achieved MMR in an estimated rate of 34%, CCyR was attained in rates of 53% and 58%, and PFS rates were 88% and 75%, whereas those with initial MDR1/GUS ratios