2.3. Stability and Sensitivity of IG/TR Rearrangements as MRD Targets
Since IG/TR rearrangements are not directly related to the oncogenic process, they may vanish over time due to the outgrowth of subclones or ongoing and secondary rearrangements in leukemic blasts with active IG/TR recombination machinery. This might lead to an underestimation of MRD level if a rearrangement is only present in a small subclone, or even a false-negative MRD result if the rearrangement is fully lost during the disease course. It has therefore been recommended to use at least two leukemia-specific rearrangements to detect MRD in ALL to lower the risk of obtaining a false negative result
[66].
Studies comparing the rearrangement profiles at diagnosis and during the disease course or at relapse are almost exclusively focusing on pediatric patients. It has been shown that oligoclonality at diagnosis is the most powerful predictor of ongoing clonal evolution in ALL: particularly in childhood BCP-ALL, significant differences in stability were observed between monoclonal and oligoclonal rearrangements: 89% of monoclonal vs. 40% of oligoclonal rearrangements are preserved at relapse
[67]. In this study, roughly 85% of monoclonal
IGH and
TRD rearrangements remained stable between diagnosis and relapse. Among monoclonal IGK-Kde rearrangements the percentage of stable targets is even higher (95%), probably due to their end-stage character
[68]. A study comparing IG/TR profiles at diagnosis and relapse of B-ALL employing high throughput IG/TR sequencing confirmed that the overall stability of IG/TR rearrangements is rather low in (27% of clonal rearrangements were preserved), but also showed that the stability of large clones is way higher (84%)
[69]. At relapse, the general characteristics of the IG/TR gene profiles are comparable to those at diagnosis but exhibit a lower degree of oligoclonality and more frequent
TRD gene deletions, which fits with the hypothesis of ongoing clonal selection and continuing rearrangements
[67]. In T-ALL, the IG/TR rearrangements profiles at diagnosis and relapse are more stable: 97% and 86% of TR rearrangements are preserved at relapse in adult and childhood T-ALL, respectively
[70].
TRD rearrangements are the most stable ones (100% of rearrangements preserved at relapse), followed by
TRG (89%) and
TRB rearrangements (82%)
[70].
Besides different stability during the disease course, IG/TR MRD targets also vary in sensitivity of the derived real-time quantitative polymerase chain reaction (RQ-PCR) assays. The sensitivity is primarily determined by the combinatorial and junctional diversity of the CDR3 regions. Therefore, RQ-PCR assays based on rearrangements from IG/TR loci that contain more V/D/J gene segments in their germline sequence (higher combinatorial diversity) generally have higher sensitivity. Similarly, complete rearrangements that contain D-segments (
IGH,
TRB,
TRD V-D-J rearrangements) provide higher sensitivity than complete rearrangements without a D-segment (
TRG,
IGK rearrangements) and incomplete rearrangements. For example, complete
IGH rearrangements represent the most sensitive group of targets, usually reaching the sensitivity of 10
−4 [43]. Also, complete
TRB rearrangements provide decent sensitivity thanks to their extensive junctional regions
[70]. The lower combinatorial diversity in incomplete
TRB rearrangements provides an explanation for slightly lower sensitivities in this group of targets
[53]. In contrast, the sensitivity of
TRG targets is usually considerably limited (a sensitivity of at least 10
−4 is reached in less than half of the patients), owing to the restricted size of their junctional regions and the non-specific amplification of highly abundant polyclonal
TRG rearrangements in normal T-cells
[71]. Intriguingly,
TRG rearrangements contain significantly higher number of inserted nucleotides and lower number of deleted nucleotides in T-ALL than in BCP-ALL, which seems to be the most important predictor for reaching good sensitivity
[71].