Resetting the clock

In this issue of Blood, Alexander and colleagues describe the reversal of the abnormalities in adaptive immunity following ASCT for SLE. These much needed data provide mechanistic support to immunoablative therapeutic approaches in SLE.

The rationale for autologous hematopoietic stem cell transplantation (ASCT) in systemic autoimmune diseases, such as systemic lupus erythematosus (SLE), is based on 2 major assumptions. The first is that the immunoablative conditioning regimen will lead to deletion of autoreactive cells of the adaptive immune system and, second, that the regenerating immune system will be (more) tolerant to self-antigens; in effect, “resetting the immunologic clock” to a pre-autoimmune state.¹  Although over 100 patients with severe, treatment-resistant lupus were reported to have undergone ASCT,^2,3  there are very scarce data about the impact of ASCT on the underlying pathologic processes. The key question is whether ASCT fundamentally changes the abnormal immune response observed in SLE. Alexander et al address this question by performing a detailed phenotypic analysis of T and B lymphocytes and autoantibody responses in 7 patients before and after ASCT. At baseline, patients exhibited abnormalities characteristic of active lupus, such as lymphopenia, restricted T-cell repertoire, dominance of memory versus naive T and B cells, expansion of plasmablasts and high-titer autoantibodies. Conditioning with cyclophosphamide and rabbit antithymocyte globulin (ATG) achieved the expected lymphodepleting effect. The novelty of the paper is the careful analysis of the regenerating adaptive immune system showing the reversal of all, and normalization of most, baseline abnormalities, albeit with different kinetics. The authors confirmed the previously described normalization of the restricted T-cell repertoire by 1 year after transplantation but also provided a description of the kinetics of this normalization. They observed an initial expansion of memory T cells immediately after transplantation (driven by exogenous antigens), followed by an increased output of recent thymic emigrants starting around 6 months after transplantation that led to a diverse, normal-looking T-cell repertoire. Similarly, there was a dramatic shift in B-cell subpopulations from memory to a naive B-cell dominance after transplantation with disappearance of circulating plasmablasts, a hallmark of active lupus.⁴  Accordingly, anti-dsDNA antibodies, which correlate with disease activity in lupus and are thought to be secreted primarily by plasmablasts, disappeared in all patients. The disappearance of protective vaccine-specific antibodies suggested an effect on long-lived antibody-secreting cells, which are thought to also secrete other autoantibodies, such as antinuclear antibodies and anti-Ro/SSA and anti-La/SSB. Similar to vaccine-specific antibodies, antinuclear antibodies either disappeared or decreased significantly. Interestingly, anti-Ro/SSA and anti-La/SSB levels persisted in the 2 patients who had these antibodies at baseline, which is especially intriguing because 1 of these patients flared 18 months after transplantation. The reason for the persistence of these antibodies is unclear but may reflect the resistance of some long-lived plasma cells or a difference in the availability or presentation of various autoantigens after transplantation. The clinical significance of this observation remains to be determined. There are a few limitations to the study. First, the number of patients is relatively low, but the long follow-up and the consistency of findings among the 5 patients with lasting remissions strengthen the results. The observation that CD4⁺CD25brightFoxP3⁺regulatory T cells return to the range seen in healthy controls and inactive lupus patients is limited by the lack of pretransplantation data and functional analysis demonstrating the suppressive capacity of these cells. The demonstration of thymic regeneration of the T-cell repertoire is exciting but its applicability to older populations has yet to be determined because all patients but 1 in the study were younger than 40 years and there are some concerns that older patients may lose their thymic function and may regenerate a more restricted T-cell repertoire.⁵  How this would impact lupus is unknown. Moreover, the study focused exclusively on the adaptive immune system, and it would be important to include analysis of the innate immune system as well. For example, it would be very instructive to know what happened with the interferon signature after transplantation in patients who maintained remission and those who flared.

Despite the few limitations, this is the most comprehensive study so far in this area and together the data strongly suggest that the immunologic clock has been reset to a pre-autoimmune state after ASCT and provides mechanistic support for continued exploration of ASCT in lupus. What is not clear is if these changes are specific to this approach and whether resetting the clock is sufficient to prevent the recurrence of lupus. To address the first question, it is of utmost importance to include this type of analysis in clinical studies using other immunodepleting strategies of various intensity (eg, B-cell depletion or high-dose cyclophosphamide without HSCT) to identify the changes that are crucial for success. Only time will tell if resetting the clock has a curative potential for some. But the fact that disease-free survival was around 50% in the 2 largest published cohorts suggests that this approach may not be equally effective for all lupus patients. Therefore, it is very important to identify prognostic factors that may predict response before transplantation or the reason for nonresponse or relapse after transplantation. Long-term monitoring of the innate and adaptive immune system of patients who responded to ASCT may also identify potential targets for therapies to prevent the reemergence of autoimmunity.