Identification of the true hyperdiploid multiple myeloma subset by combining conventional karyotyping and FISH analysis

Multiple myeloma (MM) is a malignant disorder of plasma cells representing the second most common hematological malignancy. The currently accepted MM pathogenetic model includes two different types of primary events, namely chromosome translocations or chromosome number alterations resulting in hyperdiploidy. Primary translocations, often associated with hypodiploid karyotype, can be found in at least 40% of patients. Among these translocations, t(4;14) and t(14;16), found in 10–15% and 2–3% of patients, respectively, are associated with a worse outcome, and included in the revised ISS. More than half of MM cases, instead, have a hyperdiploid karyotype, characterized by trisomies involving odd chromosomes. In several studies, trisomic MM is associated with a favorable outcome. In addition to these primary events, secondary aberrations can be detected at diagnosis, or acquired following treatments, including gain of 1q (+1q) present in about 35–40% and associated with worse prognosis, 13q deletion or monosomy 13 (del13q/−13), which were initially considered a poor prognostic marker only if present at the karyotype level, and 17p deletion (17p−) found in 15–20% of MM and considered a high-risk disease marker, predictive of reduced survival. Cytogenetic abnormalities in MM can be studied by conventional karyotyping (CK) or fluorescent in situ hybridization (FISH) analysis, and more recently by arraycomparative genomic hybridization (Array-CGH) and single-nucleotide polymorphism array (SNP-array). Although karyotyping may ideally describe all chromosome aberrations of the neoplastic clone, with the exception of some cryptic translocations, the amount of proliferating cells that are required hampered the routine use of CK. For this reason, in the latest European Myeloma Network (EMN) guidelines, FISH analysis on plasma cells is recommended, and should include at least t(4;14) and 17p− abnormalities, with t(14;16) and +1q suggested as well. Hyperdiploid MM (HRD MM), instead, is not routinely investigated because multiple probes are required over the conventional markers. However, in the era of novel agents, the identification of hyperdiploidy was proved to be helpful in MM prognostication since trisomic MM can benefit the most from lenalidomide treatment, and retains a favorable outcome. Starting from CK, our study aims at identifying distinctive genetic features of hyperdiploid MM that are associated with a favorable outcome, pursuing the goal of their inclusion in the routine FISH assessment with a step-by-step approach. Bone marrow of 292 newly diagnosed MM patients was studied by both CK and FISH on separated plasma cells for detecting high-risk (HR) cytogenetic aberrations, including t(4;14), t(14;16), 17p−, and +1q. Patient’s characteristics, including ISS stage, symptoms at diagnosis, type of treatment, and OS, were collected (Supplementary). Median follow-up of the study population was 42 months. Among the entire cohort, 76 (26%) patients showed an abnormal karyotype, and were selected for further analysis, while the remnant 216 cases were not evaluable or not informative. Based on CK nomenclature, karyotypes were classified into hyperdiploid (HRD, 47–57 chromosomes)

In addition to these primary events, secondary aberrations can be detected at diagnosis, or acquired following treatments, including gain of 1q (+1q) present in about 35-40% and associated with worse prognosis 6 , 13q deletion or monosomy 13 (del13q/−13), which were initially considered a poor prognostic marker only if present at the karyotype level 7 , and 17p deletion (17p−) found in 15-20% of MM and considered a high-risk disease marker, predictive of reduced survival [2][3][4]6,8 .
Cytogenetic abnormalities in MM can be studied by conventional karyotyping (CK) or fluorescent in situ hybridization (FISH) analysis, and more recently by arraycomparative genomic hybridization (Array-CGH) and single-nucleotide polymorphism array (SNP-array). Although karyotyping may ideally describe all chromosome aberrations of the neoplastic clone, with the exception of some cryptic translocations, the amount of proliferating cells that are required hampered the routine use of CK. For this reason, in the latest European Myeloma Network (EMN) guidelines, FISH analysis on plasma cells is recommended, and should include at least t(4;14) and 17p− abnormalities, with t(14;16) and +1q suggested as well 9 . Hyperdiploid MM (HRD MM), instead, is not routinely investigated because multiple probes are required over the conventional markers. However, in the era of novel agents, the identification of hyperdiploidy was proved to be helpful in MM prognostication since trisomic MM can benefit the most from lenalidomide treatment, and retains a favorable outcome 3,5 .
Starting from CK, our study aims at identifying distinctive genetic features of hyperdiploid MM that are associated with a favorable outcome, pursuing the goal of their inclusion in the routine FISH assessment with a step-by-step approach.
Bone marrow of 292 newly diagnosed MM patients was studied by both CK and FISH on separated plasma cells for detecting high-risk (HR) cytogenetic aberrations, including t(4;14), t(14;16), 17p−, and +1q. Patient's characteristics, including ISS stage, symptoms at diagnosis, type of treatment, and OS, were collected (Supplementary). Median follow-up of the study population was 42 months.
In our cohort, HRD MM OS was not significantly different with respect to HD MM, despite an increased presence of HR genetic and clinical features in HD MM patients. This can be only partially related to the not negligible presence in HRD MM patients of HR alterations, like those detected by FISH. It is also possible that a cytogenetic clone unintentionally selects a more aggressive form of MM, therefore making the difference elusive between the two groups 11,12 . In fact, the possibility of identifying an abnormal clone by CK has been related to a high mitotic rate and high percentage of bone marrow plasma cells, these variables also being correlated with the percentage of abnormal metaphases. Alternatively, it is also possible that the development of chromosome abnormalities in the malignant plasma cell may lead to a more aggressive tumor cell growth. Despite the limitations of the study (retrospective nature, small sample size, and heterogeneous treatments received), in our cohort, it clearly appears that cytogenetically defined HRD MM represents a heterogeneous group of MM where numerical changes are coupled with structural aberrations. The same evidence was reported by other groups that demonstrated that trisomies and high-risk cytogenetic alterations could coexist 5,13,14 , although with conflicting results on the outcome. Consequently, it is evident that the number of chromosomes or the ploidy level could not be enough to define the HRD MM, but rather the whole pattern of chromosome aberrations is needed to identify myeloma with hyperdiploidy. Indeed, in our cohort, we confirmed that the type and number of trisomies detected by CK have a relevance since concomitant 9/11/15 trisomies correlate to a better outcome, and represent an independent prognostic factor together with the absence of IGH rearrangement. Moreover, this association was particularly evident in T-HRD patients, where trisomies correlate with low frequencies (≤2) of FISH alterations and HR FISH.
Although at least half of MM patients belong to the HRD subset, the latest EMN guidelines do not recommend the assessment of hyperdiploid status at diagnosis 9 . The major issue of evaluating hyperdiploidy by FISH is the need of multiple probes for odd chromosomes that increases the costs, the effort of testing, and the amount of plasma cells required. Indeed, our results suggest to restrict the investigation of hyperdiploidy to those cases that in the beginning are negative for HR FISH, and simultaneously have ≤2 FISH abnormalities and no IGH rearrangement. This approach would save resources, and at the same time, would make the characterization of most of not high-risk samples possible.
The detection of ploidy status is an emerging issue, especially for HRD MM, and in a recent paper, Sidana et al. developed a flow cytometry approach based on the DNA index. With this method, high-hyperdiploidy patients had an improved OS as compared with those with low-hyperdiploidy 15 , consistent with our CK-based results.
In conclusion, the identification of T-HRD MM represents a new challenge within the heterogeneous group of HRD patients, with relevant prognostic implications. The combination of banding analysis and FISH for HR aberrations contributes to better define the complexity of HRD MM. Finally, the concomitant presence of trisomies of chromosomes 9/11/15, after exclusion of HR features, is a surrogate marker of true hyperdiploid myeloma.