Among the stalwarts of treatment for multiple myeloma are the proteasome inhibitors (PIs), particularly bortezomib. However, development of drug resistance is inevitable, predicatively accompanied by its dreaded negative impact on clinical outcome. The primary function of proteasomes is maintenance of cellular homeostasis through enzymatic degradation of unneeded or damaged proteins. Proteasomes are multi-protein complexes, located in both the cytoplasm and the nucleus, that are often depicted schematically as having a cylindrical or barrel shape. Proteasomes are composed of α- and β-subunits. The active sites of protein degradation (mediated by chymotrypsin-like, trypsin-like, and post-glutamyl peptide hydrolyzing activities) are located in the β-subunits (designated proteasome β, PSMB) facing the inside of the barrel. The α-subunits, located on the top and bottom of the proteasome, detect ubiquitinated proteins targeted for degradation, thereby serving as the gatekeepers for entry into barrel. Bortezomib inhibits hydrolysis of ubiquitinated proteins as a consequence of binding to the β5 subunit. Direct and indirect interactions with the β1 and β6 subunits, respectively, may also contribute to the proteasome inhibitory activity of bortezomib.
When in the course of treatment patients become PI-resistant is variable. One hypothesis to explain this variation is that those who develop resistance early in the course of treatment have existing genetic sequence variants (polymorphisms or mutations) that mediate the resistance while those that develop resistance later acquire the causative sequence variants through somatic mutation. In support of this hypothesis, studies have shown that PSMB5 sequence variants can arise when tumor cell lines are cultured with bortezomib. The purpose of the current study, led by David Lichter from Millennium Pharmaceuticals in Cambridge, MA, was to determine if sequence variants in PSMB genes might account for drug resistance or have an impact on clinical outcome in patients with myeloma. To investigate this hypothesis, Lichter and colleagues sequenced the coding regions of PSMB genes in pre- and post-treatment samples from patients who participated in the phase III Assessment of Proteasome Inhibition for Extending Remissions (APEX) trial of single-agent bortezomib compared with high-dose dexamethasone for treatment of relapsed myeloma. Twelve new PSMB sequence variants were identified, but no association was found between the frequency of these PSMB single nucleotide polymorphisms (SNPs) and clinical response to either bortezomib or dexamethasone nor was overall survival or time to disease progression found to correlate with PSMB SNP allelic frequency. Allelic and genotype frequency of non-synonymous SNPs (those sequence variants that result in an amino acid change in the protein product of the gene) in pre- and post-treatment multiple myeloma samples did not differ significantly from the frequency of those SNPs in the general population and no unique non-synonymous substitutions were observed in the post-treatment samples. Moreover, sequence variants occurred at similar frequencies in pre- and post-treatment samples. Together, these results suggest that treatment with bortezomib (or dexamethasone) does not exert a selection pressure that favors outgrowth of myeloma cells with mutant PSMBs and that the observed new sequence variants represent germline polymorphisms rather than somatic mutations. Notably, the PSMB sequence variants that were identified in preclinical models of bortezomib resistance were not detected in patient tumor samples collected after clinical relapse following bortezomib therapy. Thus, alternative mechanisms must underlie bortezomib resistance.
In Brief
While the outcome of this study can be viewed as negative (i.e., a mechanism for bortezomib resistance was not identified), the rigorous experimental design, the use of primary tissue samples from well-characterized patients enrolled in a phase III clinical study, and the relatively large number of samples analyzed make the results compelling. Thus, sequence variants that directly affect the structure/function of PSMBs can be confidently excluded as the cause of bortezomib resistance, and the search for other mechanisms for loss of sensitivity to PIs can proceed. Just as the mechanism for resistance to JAK2 inhibitors (heterodimerization of JAK2 with other JAK family members) differs from the mechanism that underlies resistance to the tyrosine kinase inhibitors used to treat CML (somatic affecting mutations of the ATP binding and solvent pocket of BCR-ABL), we can imagine that the basis of bortezomib resistance will be novel and that elucidating this mechanism will provide both fascinating new insights into the sinister but elegant nature of tumor biology and an empiric approach to pharmacologically subverting the drug resistance.
Competing Interests
Dr. Leleu indicated no relevant conflicts of interest.
