AUTHORS
David E. Mery, PhD, Guido Tricot, MD/PhD2, Jieqiong Zhou, MD3, Visanu Wanchai, PhD4, Eric Siegel, MS5, Cody Ashby, PhD6, Yan Cheng, PhD7, Hannah Cutshall, MD8, Christopher Wardell, PhD9, Clyde Bailey, BS10, Daisy Alapat, MD11, Hongwei Xu, MD12, Carolina D Schinke, MD13, Maurizio Zangari, MD14, Sharmilan Thanendrarajan, MD15, Frits van Rhee, MD, PhD16, John D Shaughnessy, Jr, PhD17, Fenghuang Zhan, MD/PhD18
AFFILIATIONS
1 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
2 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
3 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
4 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
5 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
6 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
7 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
8 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
9 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
10 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
11 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
12 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
13 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
14 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
15 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
16 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
17 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
18 Central Arkansas Veterans Healthcare System, Little Rock, United States, University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
ABSTRACT
Background:
Risk stratification in newly diagnosed multiple myeloma (NDMM) primarily identifies high-risk disease, whereas favorable subgroups within standard-risk disease remain poorly defined. Identifying patients likely to achieve exceptionally durable remission requires prolonged clinical follow-up, which is limited in most clinical trials. We sought to uncover tumor-cell and microenvironmental features within GEP70-standard-risk NDMM to identify a subgroup with a high probability of durable remission (> 10 years).
Methods:
Patients with NDMM treated on UAMS Total Therapy 2–5 trials between 1998 and 2016 (n = 1,041) who had baseline CD138-selected gene expression profiling (GEP) and long-term follow-up (>15 years) were evaluated. Discovery analyses were restricted to GEP70-standard-risk (Shaughnessy et al. Blood 2007: 2276) disease (84.5%). Transcriptome-wide Cox regression identified plasma-cell genes associated with time to progression at FDR <0.01, followed by consensus clustering. Following consensus clustering, cluster-associated genes were annotated by chromosome arm and biological function. Enrichment of dominant chromosomal and gene-family patterns was quantified using Fisher exact tests, with Benjamini–Hochberg correction for chromosome-arm comparisons. Paired CD138-selected and whole-bone-marrow profiles from MGUS and NDMM were compared, and the resulting risk framework was evaluated across treatment eras and against a clinically accessible classification using FISH-defined 1q copy number and uninvolved immunoglobulin measurements.
Results:
Within GEP70-standard-risk disease, transcriptome-wide Cox regression identified 211 progression-associated plasma-cell genes that defined four stable consensus clusters. Of these, 53 (25.6%) mapped to chromosome 1q, and 52 of 53 were assigned to Cluster 4, representing a 2.42-fold enrichment (OR: 192.6; FDR<0.0001). A second dominant program emerged in Cluster 1, which was strongly enriched for genes of the immunoglobulin-locus-bearing arms 22q (OR: 23.0; FDR<0.0001) and 14q (OR: 11.0; FDR=0.025). All 19 immunoglobulin genes, 12 on 22q, 6 on 14q, and 1 on 2p, were assigned to Cluster 1, representing a 5.75-fold enrichment (Fisher exact P<0.0001). To anchor these transcriptional programs to measurable disease biology, the 1q signal was modeled against weighted FISH-defined 1q21/CKS1B copy number (CN), yielding a 36-gene elastic-net GEP1q classifier that separated 2-copies (GEP1q-CN2) from ≥3-copies (GEP1q-CN≥3) disease. Immunoglobulin genes (n=31), the expression of which correlated with preservation of uninvolved serum immunoglobulins (unIGH) (Pearson r > 0.40), were combined into the PolyIG score. An optimal cut-point of 11 classified patients as PolyIG-high or PolyIG-low. Both scores were validated in independent held-out cohorts. Integrating tumor-intrinsic 1q burden with preservation of the polyclonal immunoglobulin compartment defined three biologically distinct risk groups: favorable GEP1q-CN2/PolyIG-high, intermediate GEP1q-CN2/PolyIG-low, and adverse GEP1q-CN≥3 disease. Corresponding 10-year progression risks were 19.0%, 39.0%, and 63.6%, respectively (P<0.0001). The favorable group also showed an almost intact, MGUS-like bone marrow microenvironment and retained favorable outcomes across treatment eras, with marked improvement after the introduction of proteasome inhibitors. This progression hierarchy could be reproduced clinically very well using FISH-defined 1q copy number and unIGH preservation: FISH 1q-CN2 with preserved unIGH defined favorable disease, FISH 1q-CN2 with suppressed unIGH defined intermediate disease, and FISH 1q-CN≥3 defined adverse disease, with corresponding 10-year progression risks of 21.1%, 37.4%, and 62.8%, respectively (P<0.0001).
Conclusion:
Exceptionally durable remission (> 10 years) in GEP70-standard-risk NDMM was associated with GEP1q-CN2 and high PolyIG expression, reflecting favorable tumor-cell and MGUS-like microenvironmental states. Universally available FISH and uninvolved immunoglobulin measurements represented an excellent substitute, supporting prospective risk-adapted studies.
DISCLOSURES
David E. Mery, PhD: No relevant conflicts of interest to declare.