Bosutinib is approved for newly diagnosed Philadelphia chromosome-positive (Ph+) chronic phase (CP) chronic myeloid leukemia (CML) and for Ph+ CP, accelerated (AP), or blast (BP) phase CML after prior treatment with tyrosine kinase inhibitors (TKIs). In the ongoing phase 4 BYOND study (NCT02228382), 163 CML patients resistant/intolerant to prior TKIs (n = 156 Ph+ CP CML, n = 4 Ph+ AP CML, n = 3 Ph-negative/BCR-ABL1+ CML) received bosutinib 500 mg once daily (starting dose). As of ≥1 year after last enrolled patient (median treatment duration 23.7 months), 56.4% of Ph+ CP CML patients remained on bosutinib. Primary endpoint of cumulative confirmed major cytogenetic response (MCyR) rate by 1 year was 75.8% in Ph+ CP CML patients after one or two prior TKIs and 62.2% after three prior TKIs. Cumulative complete cytogenetic response (CCyR) and major molecular response (MMR) rates by 1 year were 80.6% and 70.5%, respectively, in Ph+ CP CML patients overall. No patient progressed to AP/BP on treatment. Across all patients, the most common treatment-emergent adverse events were diarrhea (87.7%), nausea (39.9%), and vomiting (32.5%). The majority of patients had confirmed MCyR by 1 year and MMR by 1 year, further supporting bosutinib use for Ph+ CP CML patients resistant/intolerant to prior TKIs. Additional BYOND Study Investigators are listed below Acknowledgements.

Introduction: Inotuzumab ozogamicin (InO), is a novel anti-CD22 antibody-calicheamicin conjugate approved in R/R B-ALL due to its high hematologic remission rate (81%) based on the phase 3 INO-VATE trial comparing to investigators choice (IC). The TOWER trial demonstrated the efficacy and safety of blinatumomab (Blina) for treatment of Ph- B-ALL versus IC. The relative effectiveness of InO versus Blina was investigated by applying indirect treatment comparison (ITC) methods. A UK-based cost-effectiveness model (CEM) submitted to the Scottish Medicines Consortium (SMC) explored the impact of treatment differences with regard to mean life years (LY) gained and quality-adjusted life years (QALY). Methods: As R/R ALL is a terminal disease if left untreated, achievement of complete response/complete response with incomplete count recovery (CR/CRi) in conjunction with stem cell transplant (SCT) is essential for long-term survival. The three most important outcomes related to treatment are thus the level of response determined by CR/CRi, the rate of SCT, and overall survival (OS). Without potentially curative therapy such as SCT, there is no evidence to suggest long-term survival is possible. Therefore, to compare InO to Blina, comparisons of these outcomes were explored using patient-level data from the INO-VATE ALL trial and aggregate data from the TOWER trial. The CEM structure contained four health states categorising patients based on 'No CR/CRi & no SCT', 'CR/CRi and no SCT' and patients receiving SCT ('SCT/Post SCT') - with progression-free survival (PFS) and OS modelled within these states. States were clinically validated as relevant to treatment of the disease. Death was the fourth health state. Different methods were incorporated to allocate Blina patients to the respective health-states. For levels of response (CR/CRi) and SCT a matching-adjusted indirect comparison (MAIC) and a Bucher ITC were explored. As CR/CRi and SCT rates are not mutually exclusive, a multinomial ITC was also conducted. Once allocated into respective health states, OS and PFS were modelled. Three ITC methods were used to compare OS; a simulated treatment comparison (STC), MAIC and a standard network meta-analysis. In the absence of PFS data for Blina, PFS was assumed to have the same relative treatment effect as OS. Quality of life data within the model for the 'No CR/CRi & no SCT' and 'CR/CRi and no SCT' were informed from InO trial data, while SCT quality of life was informed from the literature with time-varying utilities. Costs were incorporated from a UK perspective using 2017 sources and were those submitted to the SMC. Results were annually discounted at 3.5%. Results: Health state proportions for Ph- InO patients were used as the basis to estimate corresponding Blina proportions and show 49.3% of patients treated with InO reach SCT. With higher odds for CR/CRi and SCT for InO, the ITC results consistently indicate Blina leads to lower proportions of patients receiving SCT (19.1-22.5%) and CR/CRi (25.2-33.3%). ITCs comparing OS outcomes for InO versus blinatumomab show negligible differences between treatments, consistently across the three methods. All combinations of the various methods were explored using the list price for both treatments. The results of the CEM ranged from 0.91-1.14 incremental QALYs for InO versus Blina, while LYs ranged from 2.03-2.59 resulting from higher rates of SCT. The incremental cost-effectiveness ratio (ICER) ranged from £3,700 to £7,010 for InO versus Blina. Extensive scenario analysis indicates that InO is a cost-effective option compared to Blina at a willingness to pay threshold of £20,000 per QALY. The SMC recommended InO as a cost-effective use of resources citing an ICER of £6,754 in the CEM when using the MAIC; InO was associated with a mean survival gain of >29 months over Blina corresponding to this ICER. Conclusions: Outcomes from the ITC indicate that InO provides patients with a greater probability of achieving CR/CRi and/or receiving a subsequent SCT versus Blina. As CR/CRi followed by SCT are essential for long-term survival and potential cure, the mean OS gain in the model cited in the SMC recommendation is intuitive as it aligns with the superior CR/CRi and SCT odds ratios associated with InO. Further research is required to determine the long-term PFS and OS following SCT in R/R B-ALL, beyond what can be reliably captured within clinical trials. Disclosures Critchlow: BresMed Health Solutions Ltd.: Consultancy. Cooper:BresMed Health Solutions Ltd.: Consultancy. van Oostrum:Ingress Health: Employment; Pfizer: Consultancy; Merck: Consultancy; Janssen: Consultancy; AstraZeneca: Consultancy. Welch:Pfizer Inc: Employment, Equity Ownership. Russell-Smith:Pfizer: Employment, Equity Ownership.

Aim: De novo relapsed and/or refractory acute myeloid leukemia (rrAML) has limited treatment options for patients not eligible (‘unfit’) to receive intensive chemotherapy-based interventions. The authors aimed to summarize outcomes for licensed therapies in this setting. Materials & methods: A systematic literature review identified licensed therapies in this setting. A feasibility assessment was made to conduct a network meta-analysis to evaluate comparative efficacy. Results: Seven unique trials were identified. Median survival months were 13.8 for gemtuzumab ozogamicin (GO), 9.3 for gilteritinib (FLT3 mutated rrAML), 5.6 for low-dose cytarabine and 3.2 for best supportive care; transplant rates with gilteritinib and GO were 25.5% and 19%, respectively. A network meta-analysis was not feasible. Conclusion: There remains a high unmet need in de novo rrAML patients not eligible for intensive therapy, with GO and gilteritinib (only FLT3 mutated AML) providing the best current options.

Background and ObjectiveThe phase III ALFA-0701 study demonstrated the efficacy and safety of gemtuzumab ozogamicin (GO) versus standard of care (SOC) chemotherapy (daunorubicin and cytarabine) for the treatment of adult patients with de novo CD33+ acute myeloid leukaemia (AML). This study analysed the cost-effectiveness of GO from the perspective of the UK health care payer. Methods A cohort state-transition model was developed to estimate direct health care costs and quality-adjusted life-years (QALYs) over a lifetime time horizon from AML diagnosis to death using monthly cycles. Data on complete remission, overall survival, relapse-free survival (RFS), haematopoietic stem-cell transplantation, and adverse events for GO plus SOC versus SOC were obtained from the ALFA-0701 study. Overall survival and RFS were extrapolated beyond the trial horizon using mixture cure models. Unit costs were obtained from standard national sources. Utilities were identified in a systematic literature review. Costs and outcomes were discounted at 3.5%. Analyses were performed for the base-case population, excluding patients with an unfavourable cytogenetic profile, and the overall population. Results For the base-case and overall populations respectively, incremental per-patient costs (£13,456 and £14,773) and QALYs (0.99 and 0.68) for GO plus SOC versus SOC resulted in incremental cost-effectiveness ratios (ICERs) of £13,561 and £21,819 per QALY gained. The mean probabilistic ICERs were £14,217 and £23,245, respectively. Univariate sensitivity analyses supported the robustness of the results. Conclusions The ICERs for both populations met NICE's £20,000-£30,000 willingness-to-pay threshold for medicines and supported the current approval for GO.

No abstract