Post-allograft relapse of acute lymphoblastic leukemia: rational use of the new targeted therapies

Crotta and colleagues present outcome data in this issue on a large series of pediatric and young adult patients with post-allograft relapse of acute lymphoblastic leukemia (ALL)¹. In an unprecedented era of change, the poor prognosis of this group is likely to improve with the availability of biological and immunotherapies. We consider in this editorial the impact of this study and the questions and controversies it highlights for clinical care in 2017.

The Novartis sponsored study presented by Crotta et al. uses retrospective data from the CIBMTR registry to examine outcomes for 1349 patients with relapsed/refractory (R/R) B-cell precursor ALL aged 3–21 and treated with allogeneic stem cell transplant between 2009 and 2013. Twenty-nine percent (394) of patients relapsed. This represents a poor prognosis group in contemporaneous parallel studies² and smaller historical cohorts of pediatric and young adult ALL^3,4. Kaplan–Meier estimated survival probability post relapse was 38.7% at one year and 20.2% at five years, higher than reported for adult relapsed ALL in retrospective European registry studies where five-year survivals less than 10% are reported⁵. This is likely to be due to the poorer outlook of adult patients but also raises the question of whether any patients in this study received chimeric antigen receptor T-cells (CARs) or any of the other novel targeted therapies. The study acts as a benchmark in providing baseline outcome data and provides important evidence for the efficacy of second allograft (five-year survival 21.6%).

In 2017 the optimal management of post-transplant relapse remains unclear. In less than a decade the outlook for these patients has changed from one where discussion of options would be a choice between best supportive care, palliative chemotherapy or re-induction chemotherapy with the potential goal of second allograft to a changing landscape of novel therapies, namely blinatumomab, inotuzumab and CARs.

Blinatumomab, a BiTE (bispecific T-cell engaging) monoclonal antibody directed to leukemic CD19 and host cytotoxic T-cell CD3, is an effective agent in relapsed ALL. The Amgen sponsored TOWER trial recruited adults with relapsed/refractoryPhiladelphia chromosome negative (Ph-) B cell precursor ALL (Central Nervous System (CNS) negative) including patients who had relapsedafter an allograft (18%). Overall complete remission (CR) ratesof 34% versus 16% for standard of care (SOC) chemotherapywere demonstrated as well as a survival benefit (medianoverall survival (OS) 7.7 months versus 4.0 months⁶. Patients in the blinatumomab group were more likely to achieve minimal residual disease (MRD) negativity than the chemotherapy comparator arm (76% versus 48% of those patients achieving remission). Comparable remission rates have been documented in pediatric patients in phase 1 trials⁷ and in Philadelphia chromosome positive (Ph+) adult patients⁸. Significant adverse events are neurological toxicities and cytokine release syndrome. Inotuzumab ozogamicin an anti-CD22-calecheamicin conjugate also led to impressive CR rates (80.7% versus 29.4% for standard intensive chemotherapy) given for up to 6 cycles in the phase 3 INO-VATE trial for adult patients with relapsed/refractory acute lymphoblastic leukemia including Ph + ALL. Overall survival difference did not achieve significance (7.7 months versus 6.7 months) and veno-occlusive disease (VOD) was a significant concern. However, restricted mean survival time was significantly longer in the Inotuzumab treated arm’ (13.9 vs. 9.9 months) and longer term follow up shows 23% two-year survival. Initial experience in pediatric patients confirms a similar toxicity profile⁹ and CR rates of 62% in heavily treated patients based on published abstract data¹⁰. Further studies specifically addressing the hepatic adverse event profile of inotuzumab confirm that prior Hematopoietic Stem Cell Transplant (HSCT)’ (particularly where alkylating agent chemotherapy is used) and pre-existing liver abnormalities are a significant risk factor for the development of veno-occlusive disease¹¹.

The recent FDA approval of Novartis CAR T-cell therapy CTL019 for the treatment of relapsed (second or later relapse/refractory B cell precursor ALL in children and young adults) follows the landmark ELIANA study (NCT02435849), a single arm open-label multi-center global phase 2 study for pediatric/young adult patients (age up to 21 years) with CD19+ relapsed/refractory ALL. This study followed on from a single center study of CTL019 showing CR rates of 90% with durable responses and CAR T-cell persistence at 6 months¹². ELIANA has reported in abstract form¹³ outcomes for 68 patients infused with a single dose of CTL019 following T-cell depleting chemotherapy with overall survival probability of 79% at 12 months (median follow up 6.4 months) for patients infused. The requirement for a leukapheresis procedure and laboratory manufacturing process adds a time constraint and this is reflected in the fact that not all eligible patients are infused (manufacturing failure rate 8% and 10% not infused due to death or adverse events). Cytokine release syndrome (CRS) occurs frequently after CAR T-cell infusion secondary to T-cell activation and proliferation and in severe cases causes vascular leak syndrome and potential need for ITU admission. CRS occurred in 78% of patients (48% grade 3 or 4) in ELIANA although remarkably there were no CRS related deaths¹⁴. Allogeneic CARs may represent a solution to manufacturing issues and access for CAR T-cells if similar efficacy and safety are demonstrated.

The ground-breaking efficacy of novel agents and CAR T-cells comes at the price of toxicities as outlined above. As these therapies become embedded into mainstream clinical practice experience grows in toxicity management and how we might modify other therapies accordingly. Steroids are used pre-emptively and as a treatment for CRS and neurological toxicity with blinatumomab but significant toxicities will limit its administration. Strategies to manage the liver toxicities of inotuzumab are yet to be evaluated but avoidance of co-administration of hepatotoxic drugs, careful selection of patients for inotuzumab based on prior therapy given and limitation of the number of cycles of drug given are all potential strategies. Steroids and tocilizumab are usually given for clinically significant cytokine release syndrome with CAR T-cells and CTL019 cells do continue to expand and persist after tociluzumab use¹⁵. Prediction of who will develop severe CRS through development of biomarkers and regression modeling is the subject of ongoing research¹⁶ and the potential to use either inotuzumab or blinatumomab followed by CARs in a sequential fashion to enable patients to proceed to CARs with low disease burden makes severe CRS and neurologic toxicity less likely in this scenario.

There remain many unanswered questions when faced with the difficult scenario of post-transplant relapse. Having set goals for therapy, management should be individualized and requires assessment of disease burden, co-existent graft versus host disease (GVHD) and co-existent active CNS involvement as well as consideration of the toxicity profile of each available agent to each patient and any relevant quality of life concerns. Logistical issues and disease progression whilst awaiting treatment may also govern options. Not all therapies will be available to all; however, it seems reasonable to attempt to achieve a deep remission in most patients along with retention of good performance status. Novel agents may represent the key to drive MRD negative deep remissions. It remains to be determined whether this will reduce the risk of relapse with a subsequent second allograft. The role of a second allograft following an MRD negative remission achieved with these agents is also unclear although it is clear that a substantial minority become long term survivors. CAR T-cell therapy is an alternative in good responders and probably a better option in those with less deep remissions.

CAR T-cell therapy is still in its infancy. The etiology of relapse after CTL09 is an area of expanding research interest – in the majority of cases this is associated with either lack of persistence or CD19 negative escape. High expansion and long term persistence with CTL019 > 1 year is associated with durable responses¹⁵ but how long is CAR T-cell persistence required to achieve long term cure? CD19 negative escape once again demonstrates the plasticity of leukemia and its ability to undergo clonal diversification and evolution, one of the primary problems underlying successful targeted therapeutics¹⁷. Lineage switch has been reported with global lineage reprogramming¹⁸. Can dual targeting of CD19 and CD123 provide a solution¹⁹? Building on initial successes, the future of CAR therapy is likely to lie in achieving long term persistence and minimizing toxicity as well as a better understanding of the mechanisms underlying immune escape.

The therapy of ALL is rapidly evolving. Having demonstrated the efficacy of the targeted therapies in relapse, the next task is to evaluate them earlier in the disease course, in the hope of reducing the use of CR1 allografts. CAR T-cells will also likely be used at earlier stages, particularly if we can overcome issues of logistics, cost and availability.

Transparency

Declaration of funding

This editorial received no funding.

Declaration of financial/other relationships

C.L.F. has disclosed that she has received travel grant and conference attendance support from Jazz Pharmaceuticals and Amgen. D.I.M. has disclosed that he does consulting work for Pfizer, Amgen and Novartis.

CMRO peer reviewers on this manuscript have no relevant financial or other relationships to disclose.