Background
Salvage therapy of relapsed/refractory (R/R) B-cell Acute Lymphoblastic Leukemia (ALL) remains an unsolved issue [
1]. Aside standard chemotherapy, which is the best treatment choice between blinatumomab, CAR-T cells and IO, that currently represent the most promising therapeutic options? [
2] To the best of our knowledge, there are no shared guidelines on the management of R/R B-ALL, even more considering the particular setting of EM disease. On one hand, despite the impressive activity demonstrated in the setting of Minimal Residual Disease (MRD) positive ALL patients, blinatumomab seems to be less effective in the context of EM involvement [
3]. On the other hand, the recent clinical trials assessing the role of IO in R/R patients did not specifically focus on EM disease. Therefore, the investigation of the role of IO in patients with such a poor prognosis is strongly required.
Discussion and conclusion
Thanks to its good safety profile and a wide experience in clinical practice, blinatumomab is currently considered the first therapeutic choice in R/R Ph- ALL [
6,
7]. However, there are at least two still-open issues about its effectiveness: the occurrence of CD19- immune escape, and its poor effect on EM disease. Firstly, loss of CD19 has been demonstrated to represent a key element of resistance [
3]. Splice variants, point mutations, lineage switch, CD81 deficiency are some of the better described CD19- relapse mechanisms [
8‐
11]. Moreover, as recently reported, a high count of T reg lymphocytes, and an impaired T reg/ T eff ratio, have been described to be predictive for blinatumomab resistance, mainly due to interleukin-10 production, resulting in suppression of T-cell proliferation and reduced CD8-mediated lysis of ALL cells [
12]. Secondly, to the best of our knowledge, studies investigating or explaining the low efficacy on EM disease are still lacking.
Furthermore, Aldoss et al. suggested that a previous history of EM disease localization before blinatumomab administration predicts a lower CR rate (
p < 0.05), EM relapse, and progression during therapy. One of the most interesting findings was that, in about 20% (5/32) of non-responders, a marrow CR with evidence of simultaneous leukemia progression in EM sites was documented [
3].
On the other hand, IO has shown similar CR and Overall Survival (OS) rates with a different safety profile [
13,
14]. A phase III randomized study (INO-VATE) demonstrated the superiority of IO compared to standard chemotherapy in terms of CR rates, underlining the potential role of the drug as a bridge-to-transplant option [
15]. IO consists of a toxin (calicheamicin) conjugated to an anti-CD22 antibody [
2], which allows its delivery to blast cells; hence, its toxicity does not seem to involve immune cells and, differently from what reported concerning blinatumomab, no cases of CD22 antigen loss have been described so far. Moreover, lineage switch mechanism (myeloid conversion), described mostly in CAR-T cell therapy, does not seem to involve CD22 expression: the antigen is maintained on the intermediate phenotype relapses, suggesting that simultaneous pressure on CD19 and CD22 might avoid this resistance mechanism [
13]. Although experimental studies have not specifically evaluated the effectiveness of IO on EM disease, its mechanism of action suggests that EM-lesion’s microenvironment and resident cell subpopulations may have a more negative influence on blinatumomab than on IO effectiveness.
These case reports, coming from our single-center experience (totally 17 patients treated with IO) described the use of IO in EM-ALL relapses in two different settings. In both cases IO was the therapeutic choice, based on CD22 positivity assessed by immunohistochemistry on EM-lesion biopsy (Figs.
1g, h,
2d-f). Both patients achieved a bone marrow CR (Case 1 MRD negativity too) and a significant resolution of the EM involvement (1 CMR and 1 PMR), allowing them to be eligible for allo-BMT.
In conclusion, despite the mild effect of immunotherapy (blinatumomab and CAR-T) on EM disease localizations has been widely described in cancer, we hypothesize that IO could represent a valid and promising therapeutic choice in this particular and dismal setting. Moreover, we suggest the assessment of CD22 status at diagnosis and relapse in all B-ALL patients, in order to better evaluate the indication for IO use. Further studies are strongly needed to confirm our hypothesis.
Acknowledgements
The authors would like to thank ELN, AIL, AIRC, PRIN, Progetto Regione-Università 2010-12 (L. Bolondi), FP7 NGS-PTL project, IMI2 HARMONY.