Novel Insight into Multi-Hit Multiple Myeloma Based on "Two-Hit" Theory of Cancer Causation

Background: Current definition of multi-hit multiple myeloma (MM) is based upon the number of high-risk cytogenetic abnormalities (CA). But we may overlook the influence of standard-risk CA and different concurrent patterns. In fact, standard-risk t(11;14) and del(13q) may bring extra danger when concurrent with other CA. And the concurrency of two secondary CA may do more harm to patients than that of one secondary CA with one primary CA. This study is to answer whether CA number or pattern exert an impact on outcomes of MM patients.

Methods: This study was carried out based on the prospective, non-randomized clinical trial BDH 2008/02. 537 MM patients with complete cytogenetic data were enrolled, of whom 64% (341/537) patients were treated with bortezomib-based three-drug induction therapy, and the remaining patients with thalidomide-based therapy. Autologous stem cell transplantation (ASCT) was recommended post induction therapy in transplant-eligible patients, and all patients received maintenance therapy for two years. CA were divided into primary CA [pCA: any type of IgH breakage], and secondary CA [sCA: del(17p), del(13q), gain(1q) (≥3 copies)]

Results: In the era of novel agents, patients with pCA only did not have outcomes different from those patients without any FISH abnormality. Patients with s1 CA or s1+p CA had hazard ratio for PFS or OS of 1.5-2.0. Patients with s2 CA or s2+p CA had hazard ratio for PFS or OS of 2.0-3.0. Patients with concurrent del(13q), del(17p) and gain(1q) (s3 CA) had hazard ratio for PFS of 3.11 and for OS of 3.00. Patients with s3+p CA had hazard ratio for PFS of 4.65 and for OS of 6.16. Based on these results, we divided patients into four subgroups: no CA or only pCA, s1 CA in the presence or absence of pCA, s2 CA in the presence or absence of pCA, and s3 CA in the presence or absence of pCA. Both the PFS and OS decreased in a stepwise fashion with the accumulation of CA (Figure 1). Therefore, we defined double-hit MM as any one sCA in the presence or absence of pCA. Triple-hit MM referred to two sCA plus pCA or not, and quadra-hit MM referred to three sCA plus pCA or not. Furthermore, we confirmed the prognostic independence of CA pattern in the multivariant analysis with International Staging System (ISS) and LDH (Table 1).

Conclusion: In this study, we found that the primary CA as a whole lost its adverse effect when treated by bortezomib-based regimens. CA subtype conferred a prognostic value. In details, secondary CA imposed an accumulative risk to patients. For the first time, we indicated that double-hit or triple-hit MM should be defined upon the number of secondary CA. This definition coincides with the “Two-Hit” theory of cancer causation and fits well with MM evolution model. The prognostic significance of CA pattern needs validation in further prospective trials.