Analysis of cereblon-binding protein expression in patients with newly diagnosed multiple myeloma

Lenalidomide, a member of the immunomodulatory drug (IMID) family, is routinely used in the treatment of both newly diagnosed and relapsed refractory (RR) multiple myeloma (MM) and is highly effective when combined with dexamethasone.¹ By binding to cereblon (CRBN), lenalidomide selectively targets the degradation of Ikaros family zinc finger protein (IKZF) 1 and IKZF3 which ultimately leads to MM cell death. The importance of IMID-induced degradation of IKZF1 and IKZF3 has been shown in vitro, but it is unclear if their expression levels are of prognostic value. Another CRBN-binding protein identified by Zhu et al. as being of prognostic relevance is karyopherin subunit alpha (KPNA) 2.²

Previous studies have shown conflicting results when investigating the importance of CRBN-binding protein expression levels in patients with newly diagnosed or IMID-treated RR MM. Krönke et al. identified an association between high IKZF1 mRNA expression and inferior progression free survival (PFS) and overall survival (OS).³ This contrasts with results published by Zhu et al. showing that low IKZF1 expression levels (using gene expression profiling) are an adverse prognostic factor for outcome in patients with RR MM treated with an IMID-based therapy.² Furthermore, Sehgal et al. did not identify any correlation between baseline IKZF1/IKZF3 levels and response to IMID treatment or survival rates.⁴

This study by Katharina Kriegsmann, Heidelberg University, and colleagues aimed to confirm if the expression levels of the CRBN-binding proteins IKZF1, IKZF3 and KPNA2 in MM cells correlated with measurable clinical and prognostic characteristics in a large cohort of 214 patients at the time of diagnosis.⁵

Patient population and study design

• N = 214 (129 vs 85, male vs female)
• Patients had newly diagnosed MM prior to induction therapy with lenalidomide and were randomized to participate in the phase III GMMG-HD6 trial (NCT02495922)
• Median age: 60 (37–70) years
• ISS stage (I vs II vs III): 80 (37.4%) vs 81 (37.9%) vs 53 (24.8%)
• High-risk cytogenetic abnormalities
  • Classification according to Neben et al.⁶: deletion 17p13 and/or translocation t(4;14), gain of 1q21 (>3 copies): 40 (20.5%)
  • Classification according to Krönke et al.³: deletion 17p and/or translocation t(4;14), or t(14;16): 46 (23.6%)
• CD27 negative: 3.4% (n = 7)
• CD81-positive: 69.3% (n = 140)
• Patient age, gender, serum heavy and light chain type by immunofixation and International Staging System (ISS) status were evaluated (acquired from data collected in the GMMG-HD6 trial)
• Analysis by flow cytometry

Key findings

• No statistically significant differences in expression levels of IKZF1, IKZF3 or KPNA2 were observed when comparing patients by age, gender, light chain type, ISS type or cytogenetic risk group
• CD27 and CD81 status:
  • CD27: no statistical difference in expression when comparing CD27-positive and -negative cells
  • CD81: KPNA2 expression was increased in CD81-negative cells (1919 vs 1432, P = 0.046) – the prognostic value of this is yet to be determined
• Genetic analysis:
  • Higher expression of IKZF1, IKZF3 and KPNA2 in hyperdiploid cells with gain of chromosomes 5, 9, 15 and 19
  • Lower levels of IKZF1, IKZF3 and KPNA2 in patients with translocation t(11;14) but there was no differences in patients with other translocations involving chromosome 14
  • Patients with 19q13 had higher expression of IKZF1, IKZF3 and KPNA2
  • Patients with gain 11q22.3 had higher IKZF1 but not IKZF3 or KPNA2
  • Abnormalities del 17p13, del 13q14 and t(4;14) were not associated with increased or decreased IKZF1 or IKZF3 levels
  • Gain 1q21 was not associated with low IKZF1 expression

Conclusion

The evidence for using CRBN-binding protein expression levels as predictive factors is conflicting.^2–5 This study did not find any significant association between the expression levels of IKZF1, IKZF3 and KPNA2 when looking at age, gender, light chain type, ISS stage and cytogenetic risk. Additionally, no association between KPNA2 expression and prognostic values are reported, unlike previous reports from Zhu et al.⁵

One major finding of this study however, is that hyperdiploid MM cells with the gain of chromosomes 5, 9, 11, 15 and/or 19 showed increased expression of IKZF1, IKZF3 and KPNA2. The presence of hyperdiploid MM cells is associated with improved outcome to treatment which could possibly be explained by an increase in the expression of certain proteins.⁷ Therefore, the authors of the study hypothesise that increased CRBN-binding protein expression in hyperdiploid cells may indicate a higher dependence on the IMID pathway and so treatment with IMIDs may confer a better response. Currently, the results of studies investigating expression levels with response to lenalidomide therapy are awaited. In contrast, patients with t(11;14) showed low expression levels of IKZF1, IKZF3 and KPNA2. However, this genetic marker did not confer inferior outcome to treatment and therefore complicates the interpretation of these results.

This study has shown flow cytometry is a method that can quickly assess protein expression levels on a large scale, but notes that a standardization of protocol for measuring IKZF1, IKZF3 and KPNA2 is required. Other studies are also required to determine comparability of results across studies, where different analysis methods are used.  

In future, studies evaluating CRBN-binding protein levels after lenalidomide treatment would provide crucial information on their predictive value. However, it is difficult to obtain bone marrow samples with sufficient MM cells following induction therapy as usually good response rates are achieved. Nevertheless, the analysis of CRBN-binding protein expression at first relapse following IMID treatment will be explored.