Clonal Hematopoiesis Impacts Frailty of Newly Diagnosed Multiple Myeloma Patients: A Retrospective Multicentric Analysis

doi:10.21203/rs.3.rs-4930569/v1

Somatic mutations of hematopoietic cells in peripheral blood of normal individuals refers to clonal hematopoiesis of indeterminate potential (CHIP) and is associated with a 0.5–1% risk of progression to hematological malignancies and cardiovascular diseases. CHIP has been reported also in Multiple Myeloma (MM) patients but its biological relevance remains still to be elucidated. Here, high-depth targeted sequencing on peripheral blood derived from 76 NDMM patients revealed CHIP in 46% of them with a variant allele frequency (VAF) between ~1% and 34%: the most frequently mutated gene was DNMT3A followed by TET2. A more aggressive disease features were observed among CHIP carriers, which also exhibited more high-risk (ISS and R-ISS 3) stages than controls. Longitudinal analyses at diagnosis and during follow-up showed slight increase of VAFs (p=0.058) for epigenetic (DNMT3A, TET2, and ASXL1) and DNA repair (TP53) genes (p=0.0123); a more stable frequency was observed among other genes, thus suggesting different temporal dynamics of CH clones. Adverse clinical outcomes, in term of overall and progression-free survivals, were observed among CHIP carriers, who also exhibited immune T-cells weakening and enhanced frailty status that predicted the greater risk of toxicity and consequent shorter event-free survival of this group. Finally, a correlogram analysis identified platelets count as biomarker for higher VAF among CHIP carriers, regardless of specific variant. Overall, our study, by highlighting specific biological and clinical features, paves the way for designing tailored strategies among MM patients carrying CHIP.

Health sciences/Oncology/Cancer

Health sciences/Oncology/Cancer/Cancer genetics

Multiple Myeloma

biomarkers

CHIP

toxicity

frailty

Multiple myeloma (MM) is an incurable malignancy of plasma cells that grow within a permissive bone marrow (BM) microenvironment, supporting tumor cells transformation, proliferation and drug resistance occurrence.¹ During recent years, available therapeutic approaches have shown promising results in clinical setting of MM patients, but unfortunately it still remains incurable due to frequent relapses.² In such a context, identification of disease-specific biologic features represents a valid strategy for improving MM clinical management. Genetic-molecular alterations including deletion 17p, t(4;14), t(14;16), t(14;20), amp/gain 1q, del 1p or TP53 mutations and high-risk gene expression profiling signatures are the most robust predictors of outcomes in MM with their combination conferring an even worse prognosis.³ However, although the identification of these features is crucial for prognosis and selection of the most appropriate strategy, aging, defined as impaired organ function and reduced physiological reserves, adds greater complexity that makes patients stratification extremely challenging.⁴ During recent years, several tools have been developed for a comprehensive assessment of MM patients’ frailty; however none of available biomarkers is currently able to effectively prevent relapse, reduce mortality, and ultimately cure this blood cancer.

More recent studies based on next-generation sequencing (NGS) have shown the presence of recurrent somatic mutations in blood of healthy adults, a condition referred as clonal hematopoiesis of indeterminate potential (CHIP).⁵ Remarkably, presence of somatic mosaicism in tissues during aging, is quite ubiquitous, but when the acquired variant confers growth advantage, mutant undergoes to clonal expansion. Somatic variants influencing cell fitness may occur in every tissue, but the wide availability of blood for serial genomic studies coupled with the blood circulation and interaction with all the other tissues, makes CHIP interesting for its clinical consequences.⁶ Mutations driving the clonal expansion mostly involves leukemia-associated drivers, with DNMT3A, TET2 and ASXL1 being the most affected genes, followed by JAK2, TP53 and a large list of other genes less frequently mutated, but still attractive as study subjects for their contribution to fitness gain.⁷ Currently, definition of CHIP requires the presence of a clone with a variant allele frequency (VAF) > 2% of the molecules, reported as the least clones having a clinical relevance, in a person without a hematologic malignancy.^5,8 Nevertheless, significance of these clones, remains to be assessed with further and long termed studies.⁹ It has been demonstrated that CHIP is associated with aging, smoke and exposure to radiation and cytotoxic chemotherapy.^10–12 In addition, it is significantly associated with an increased risk of all-cause death that cannot be explained by the development of a myeloid malignancy, which incidence in elderly is less than 0,1%.^7,13,14 Among the overall causes of death, an augmented cardiovascular disease risk appears to be related to CHIP, with an increased incidence of coronary heart disease and ischemic stroke in mutation carriers.^7,15 Similarly, patients with CH are at higher risk of other inflammatory conditions including chronic obstructive pulmonary disease¹⁶ and gout,¹⁷ which are mediated by dysregulated inflammatory signaling of mutant macrophages.¹⁸ Remarkably, CHIP is also detected in cancer patients, including those with hematologic malignancies.^19–22 Indeed, CHIP has been found also among MM patients with a prevalence of 21,6% at the time of ASCT (VAF of at least 1%);^23–25 importantly, its presence was associated with shorter OS and PFS as well, particularly in those who did not receive maintenance therapy with an IMiDs. Of note, the increased mortality in patients with CHIP was not related to the increased risk of developing therapy-related myeloid neoplasms (TMN), as observed in lymphomas patients following ASCT,¹¹ nor to an increase in cardiovascular events, but it was mainly due to disease progression, possibly related to a greater risk of developing toxicity during treatment, or to an inflammatory-prone bone marrow microenvironment supporting tumor cells growth.²⁵ Although these evidences, there is still little knowledge about the clinical impact of CHIP among MM patients, mainly in those ineligible for high-dose therapies.

Here we show, through a retrospective and multicenter analysis, CHIP and its related mutations prevalence among NDMM patients, which results in more aggressive disease and poorer clinical outcomes. Notably, CHIP patients are at greater risk of developing toxicity during treatment, likely due to altered distribution of immune-cells and enhanced frailty, which anticipates shorter event-free survival.

STUDY DESIGN

A total of 76 patients diagnosed with MM according to the revised International Myeloma Working Group (IMWG) criteria,²⁶ from 2019 to September 2021, at three different hematologic Italian centers (Genoa, Catania and Cagliari) and whose peripheral blood mononuclear cells (PBMCs) were available at diagnosis and follow up for sequencing analyses were studied. Patients' characteristics are detailed in Table 1. CD138^POS and CD138^NEG cells derived from Bone Marrow (BM) aspirates were isolated with an immune-magnetic bead-based strategy (MACS system, Mylteni biotech), as previously reported.²⁷ The study was conducted under all national and international ethical and legal recommendations, following approval by the local Ethics Review Committee, in accordance to the declaration of Helsinki. (CER Liguria: 626/2022 - DB id 12752, approved at 3th July 2023). All patients gave informed consent to the study.

Table 1

Patients' characteristics.
	All cohort (n = 76)	CHIP (n = 35)	No CHIP (n = 41)	p value
Age at diagnosis
Median (range)	71 (40–90)	72 (46–84)	71 (40–90)	0.52*
< 50	4	1	3
50–59	13	6	7
60–69	17	7	10	0.9**
70–79	29	15	14
> 80	13	6	7
Gender
Male	41	19	22	1***
Female	35	16	19
Myeloma subtype
IgA kappa	9	4	5
IgA lambda	4	1	3
IgG kappa	29	15	14	0.84**
IgG lambda	19	7	12
Kappa-light chain	9	4	5
Lambda-light chain	3	2	1
Missing	3	2	1
Biochemical markers
Albumin (mg/dL)	35,9	35,69	36,09	0.461****
β-2-microglobulin (mg/L)	6,8	9,04	4,85	0.05****
Creatinine (mg/dL)	1,9	2,41	1,47	0.66****
Calcium (mmol/L)	9,44	9,53	9,37	0.075****
Hemoglobin (g/dL)	11,24	10,61	11,78	0.019****
Platelet count (x10^6)	229,3	224,9	233	0.609****
LDH (U/L)	226,1	237,6	216,3	0.335****
BM plasma cell
≥ 10%, median (1q;3q)	50 (18, 70)	55 (24, 80)	50 (15, 66)	0.2295****
≥ 60%, median (1q;3q)	75 (70, 80)	80 (70, 81)	70 (65, 80)	0.0338*
Disease stage
ISS1	13	2	11	0.0051***
ISS2	26	12	14
ISS3	29	20	9
R-ISS1	7	1	6	0.0001***
R-ISS2	29	9	20
R-ISS3	29	23	6
EMD (Y/N)	48/27	24/11	24/16	0.58*****
Induction therapy
Anti-CD38 cont. regimens	33	15	18
PIs cont. regimens	25	13	12
Lent cont. Regimens	9	4	5	0.83**
Other	9	9	6
LOT mean	1.43 (0–5)	1.53 (0–4)	1.35 (1–5)	0.1961****

*t-test

**Fisher test

***chi^2 test

****Kolmgorov-Smirnov test

*****binom. test

SAMPLE PREPARATION, SEQUENCING AND DATA ANALYSIS

DNA was isolated with QIAamp DNA Mini Kit (Qiagen) from whole peripheral blood and BM samples (using CD138^POS and CD138^NEG fractions as well), when available. Next Generation Sequencing of those samples was performed at the time of diagnosis and during follow up, by employing an Illumina Custom Enrichment panel of recurrently mutated genes in myeloid cells (N = 36). The genes list is detailed in Table 2. Libraries were generated by the Illumina® DNA Prep with Enrichment workflow, following the manufacturer’s instruction; quality and size distribution were determined using Qubit fluorimeter and Agilent TapeStation system. 2 x 150 cycles, pair-end sequencing at 500 x median coverage depth was performed through the Genomics Core at IRCCS Istituto Giannina Gaslini on MiSeq platform (Illumina). The analysis of the data was performed with the BaseSpace® Software (Illumina) and the "Dragen Enrichment" pipeline with the “somatic” setting. Intronic and synonymous variants with no impact on splicing, missense and short ins/dels reported as “benign” or likely benign” in ClinVar were excluded.²⁸ Variants not reported in COSMIC, and with CADDphred score < 25, were filtered out as well as ones reported with neutral/uncertain impact on protein function.²⁹ The Variant allele frequency (VAF) cut-off > 1% was used to define CHIP.^30,31 Discovered variants with a VAF > 40% or those with a frequency in general population < 1% were filtered out. The selected somatic mutations and their VAFs were correlated with demographic and clinical parameters, including age, sex, ISS, R-ISS stage, outcomes, and occurrence of adverse events. Subsequent analyses at 12–24 months after therapy were also run and for those MM patients with available BM samples, the tumoral and non-tumoral fractions (CD138^POS and CD138^NEG cells, respectively) were analyzed in parallel.

Table 2. Panel genes list used

Gene	Target region (exon)	Gene2	Target region (exon)3	Gene4	Target region (exon)5
ASXL1	full	GATA2	full	PPM1D	5, 6
BCOR	full	GNAS	8, 9	PTEN	full
BCORL1	full	GNB1	5-7	PTPN11	2-4, 8, 13
BRAF	11, 15	IDH1	4	RAD21	full
BRCC3	full	IDH2	4	RUNX1	full
CBL	8, 9	IKZF1	full	SF3B1	13-16
CREBBP	full	JAK2	12, 14	SMC3	full
CUX1	full	KRAS	2, 3	SRSF2	1
DNMT3A	full	MPL	10	STAG1	full
EZH2	full	MYD88	3-5	STAG2	full
FLT3	14, 15, 20	NF1	full	TET2	full
GATA1	full	NOTCH1	26-28, 34	TP53	full

MULTIPARAMETER FLOW CYTOMETRY (MFC) ANALYSIS

MFC analysis was performed at local laboratories on bone marrow (BM) samples collected at diagnosis or at any time before induction-therapy starting. EDTA blood (2 ml) was bulk lysed with 1 × BD Pharm LyseTM Lysing buffer (30 ml) for 5 min, centrifuged at 1,500 rpm for 7 min and washed once in Dulbecco's PBS. Cells (50 µl at 10–20 × 10⁶/ml) were stained for cell surface markers with 20 µl antibody combinations for 15 min at RT. Intracellular nuclear (n) and cytoplasmic (cy) staining were performed after cell fixation and permeabilization using Intrastain kit by DAKO (Milan, Italy). The following monoclonal antibodies (MoAbs) combinations were employed: 1) CD138FITC/CD56PE/CD20PerCp/CD117APC/ CD45APC-H7/CD38PE-Cy7 2) cyKappaFITC/cyLambdaPE/CD19PerCp/ CD56APC/CD45APC-H7/CD38PE-Cy7. From the first combination, we obtained plasma cells quantification; from the second combination, we evaluated plasma cells immunophenotype and clonality. Acquisition and analyses were performed using FACSCantoTM II (Becton Dickinson, Mountain View, CA), and DiVa software: a minimum of 1–2 x 10⁴ of events for each sample were acquired. As controls, anti-isotype mouse antibodies were used. Importantly, CD38bright/SSC low population was representative for plasmacell fraction; cytofluorimetric data were analyzed when an abnormal fraction of plasmacell was detected. Based on expression levels of each cluster of differentiation (CD) measured, two categories were identified: bright-expressors and low(non)-expressors; dim level was considered as latter. Samples were considered as positive when at least 20% of MM cells expressed this antigenic profile, as previously described.²⁷

STATISTICAL ANALYSIS

Data were collected in spreadsheets and were analyzed using R statistical software (v. 4.0.5; RStudio) and SPSS (v. 25; IBM). Continuous variables were expressed as mean or median and compared with Wilcoxon rank-sum or student’s t-test. Categorical variables were expressed as counts and percentages and compared using Chi-square, Kolmgorov-Smirnov, binomial or Fisher’s exact test as appropriate. Log-rank (Mantel-Cox) test was used for survival analysis between group. A P value of < 0.05 was considered statistically significant. Correlation analysis between variables was performed using Pearson’s correlation method.

Prevalence of Clonal Hematopoiesis among NDMM patients

A total of 76 MM patients whose peripheral blood (PB) samples were available at our institutions (Genoa, Catania and Cagliari) were screened for clonal hematopoiesis related mutations by using deep sequencing approaches. Table 1 summarizes demographic and clinical characteristics of the entire cohort. The median age was 71 years old (range, 40–90 years) with no gender (41 males vs. 35 females) or addictive behavior (smoking) prevalence as well; IgG was frequently observed as immunoglobulin isotype (48/76) with k and λ free lights chains occurring in 29 and 19 of these patients, respectively. The most represented induction regimens included MoAbs and PIs-based strategies with Len-based treatment used in 9 cases. Overall, next-generation sequencing (NGS) analyses revealed at least 1 CHIP variant in 46% of patients (35/76), with a median variant allele frequency (mVAF) of 0.022 (range 0.003–0.340). (Fig. 1A-B) No significant differences were observed between the CHIP and the non-CHIP carriers in terms of gender (p = 1; chi-squared test), myeloma subtype (p = 0.84; Fisher's exact test) induction regimens (p = 0.83; Fisher's exact test) and lines of therapies (p = 0.1961; binomial test). Interestingly, differently from previously reported data,²⁵ advanced age was not associated with higher CHIP prevalence although greater VAF (> 0.1) occurred among patients older than 70 years. (Fig. 1C) Nineteen patients (54.2%) had a single CHIP mutation, while six (17.1%) and ten (28.5%) had 2 or more than three mutations, respectively. (Fig. 1D) Consistent with others reports,^32–34 the most commonly mutated gene was DNMT3A (54% of cases with mVAF of 9 %) folowed by TET2 (37% of cases with mVAF of 3.0%), ASXL1 and KRAS (both with mVAF of 11%), whereas mutations in splicing factors and JAK2 were rare. (Fig. 1E and Table 3) Finally, CHIP-mutational spectrum analyses revealed majority of nonsynonymous followed by frameshift and synonymous mutations in affected genes. The gene-specific variant frequency across whole patients are summarized in Fig. 1F. Collectively, these data acknowledge CHIP as a frequently observed event among MM patients at diagnosis, in line with reported data.^25,32–34

Table 3. List of variants observed in PB of CHIP carriers at diagnosis.

CHIP is associated with greater aggressiveness and poorer clinical outcomes

CHIP negatively impacts MM patient’s clinical outcomes with shorter PFS and OS after ASCT; importantly, these negativities are cleared by IMiDs maintenance.^24,25,34 Consistent with these findings, we first sought to assess the impact of CHIP on disease progression parameters of our cohort. Initially we investigated cellular origins of these abnormalities: as shown in Fig. 2A, no significant differences were observed between CHIP and bone marrow plasma cells percentage, thus suggesting that myeloid somatic mutations are unlikely to derive from the malignant tumor cells. Indeed, the analysis of bone marrow CD138^NEG cells (i.e. fraction depleted of tumor plasma cells) showed correlation between BM and PB VAF values thus, confirming the non-tumor origin of screened myeloid mutations. (Pearson R = 0.97, p < 0.0001; Fig. 2B) Next, we examined the impact of CHIP on disease aggressiveness markers: higher β2-microglobulin (9.04 vs 4.15 mg/dl; p = 0.05), 24 h urine protein output (1.44 vs 1.10 g/24hrs; p = 0.019), creatinine (2.41 vs 1.47 mg/dL; p = 0.05) and serum FLC k (557.2 vs 250 mg/dl; p = 0.031) levels in parallel with lower eGFR (46.57 vs 64.38; p = 0.024) and hemoglobin (10.61 vs 11.78 g/dL; p = 0.019) values were found among CHIP carriers than control. (Fig. 2C) In line with these data, Mosaic plots analyses showed disease stages differences between these two groups: higher prevalence of CHIP carriers was found among high-risk patients, defined according to International Staging System (ISS) and Revised (R)-ISS staging systems score. (Fig. 2D) Together, these data confirm CHIP role as disease-aggressiveness biomarker whose evaluation could therefore improve risk stratification analysis of MM patients. Taking into account these assumptions, we next examined clinical outcomes across our cohort. As shown in Fig. 3A, C presence of CHIP was associated with significantly shorter PFS (mPFS of 493 days in those with CH vs. not reached at 5000 days in control; p < 0.0001) and OS (mOS of 925 days in CH carriers vs. not reached at 5000 days in those without CH, respectively; p = 0.025). A multivariate Cox-model CHIP focused on analysis, identified β-2 microglobulin high level as poorer clinical outcomes influencer for both PFS (HR, 1.20; 95% CI, 1.09–1.33) and OS (HR, 1.19; 95% CI, 1.05–1.34); remarkably, this parameter preserved its negative impact also among patients without CHIP. Furthermore, while platelets count significantly predicts both PFS (HR, 0.99; 95%CI, 0.98-1.00) and OS (HR, 0.98; 95%CI, 0.97–0.99) among CHIP carriers, no significant effects were observed among patients without CHIP; similarly, albumin (HR,0.85; 95%CI, 0.77–0.95) and creatinine levels (HR, 0.62; 95% CI, 0.43–0.90) significantly influence PFS of CHIP carriers.(Fig. 3B,D)

Dynamic changes of Clonal Hematopoiesis during MM progression

To investigate clonal performance over time, we used serial VAF measurements as a surrogate for clone size. Although in general we found a slight increment of VAFs over time, it was no significant. (Fig. 4A) However, by focusing on the most relevant genes for myeloid fitness, including DNMT3A, ASXL1, TET2 and TP53, we observed significant increment in their VAF during follow -up, whilst a stable frequency affected all other genes. As result these data suggest coexistence of small clones with different clonal evolution (Fig. 4B). To delineate the factors that determine each clone’s growth rate, we analyzed serial samples in 10 CHIP carriers. The median time between the first and second sample collection was 10 months (range: 3–12 months). While in general no significant increase in VAF was observed, we found that that clones bearing mutations in the fitness-related myeloid genes expanded most rapidly, with a significant increment in VAF over time. Moreover, backward tracking revealed that variants observed pre-existed at diagnosis, with either lower or similar VAF. Figure 4C shows exemplary fish-plots of two representative patients with available clinical information: lenalidomide refractory patient (MM_032) with DNMT3A and TET2 mutant clones in BM (CD138^NEG cells) and PB at diagnosis, showed increased size of both variants during treatment (starting from VAF of 2.68 and 19.26, to 2.95 and 26.9, respectively); whilst patient (MM_066) exposed to daratumumab-based regimen showed stability of DNMT3A variant after 3 months, while PTPN11 mutant, observed at diagnosis in PB, CD138^NEG and CD138^POS BM cells, resulted almost undetectable after treatment.

Collectively these data suggest that CHIP dynamics are not influenced by any specific treatment, but rather by the occurring mutant type: indeed, MDS/AML driving genes (i.e. DNMT3, TET2, TP53 e KRAS) when involved, often increase their size even during treatment, regardless of response or disease status.

CHIP presence enhances MM patients Frailty

The overall safety was not significantly different by CHIP presence although incidence of adverse events (AEs) of any grade was slightly higher in CH (97%) versus no-CHIP patients (80%); similarly, the rates of grade ≥ 3 AEs were lower in non-CHIP (36.5%) compared with CHIP carriers’ patients (85.7%). (Table 4) The most common hematological event was neutropenia of any grade, which occurred in 20 patients (57%) among CHIP group and in 13 (31.7%) among those without this abnormality. A similar trend was observed for high grade (≥ 3) hematologic toxicities, although differences were not significant between two groups (p = 0.67). Among non-hematologic toxicities, peripheral neuropathy occurred more frequently among CHIP than no-CHIP carriers, even if differences were not significant (p = 0.20 and 0.19 for any grade or higher toxicities, respectively). The most notable difference between groups was about infections, including SarSCov2 and pneumonia, which occurred in 71.4% of CHIP carrying patients and in 23% of those in control group. Importantly, most serious (AEs ≥ 3 grade) infections occurred among CHIP; unfortunately, low censured cases did not make differences significant (p = 0.36 and p = 0.37 for any grade or grade ≥ 3 toxicity, respectively). Clonal hematopoiesis has been reported to be associated with cardiovascular disease also in MM patients,³² thus we analyzed cardiovascular events rates (CVEs) in our cohort: while comparable events were observed between two groups, venous thromboembolism seemed to affect especially CHIP carriers (51.4 vs 17%), although the limited number of cases prevented significant conclusions. To support the greater infections occurrence between CHIP patient, we next screened bone marrow immune-phenotype of our cohort, where available. As shown in Fig. 5, effector T-cells resulted significantly reduced in CHIP vs. no-CHIP carriers with both CD4⁺ and CD8⁺ cells equally compromised, whilst B (CD19⁺), monocytes (CD14⁺) and NK (CD16+/CD56+/CD3-) cells were almost unaffected by CHIP presence. T-cell fitness impairment made us to calculate frailty scores of our patients by using age (0 if ≤ 75 years, 1 if 76–80 years, 2 if > 80 years), Charlson Comobidity Index (CCI) (0 if ≤ 1, 1 if > 1), and ECOG PS (0 if 0, 1 if 1, 2 if ≥ 2).³⁵ Remarkably, this approach shown frail patients’ enrichment among CHIP carriers compared with no-CHIP patients (97.1% vs 56%; p < 0.0001) thus, supporting the enhanced AEs vulnerability observed in the former. (Table 5) It is supposed that AEs occurrence leads to treatment discontinuation thus, we examined the impact of clonal hematopoiesis on event free survival (EFS) of our cohort: CHIP negatively impacted prognosis with its presence associated with shorter EFS than its absence. (Fig. 6A; p < 0.0001) A multivariate Cox-model indicates albumin level (HR, 0.92; 95%CI, 0.84-1) as the only parameter able to improve EFS among CHIP-carriers; by contrast, hemoglobin (HR, 2; 95%CI, 1.04–3.84) and β2-microblobulin (HR, 2.10; 95%CI, 1.18–3.74) values worsened outcome of patients without CHIP. (Fig. 6B) Overall, our data suggest that CHIP impairs MM patient’s fitness making them more prone to the development of therapy-related toxicities, including infection; this in turn results in greater chance of treatment delays or dose reductions that limit patient’s ability to receive optimal therapies.

Table 4. Drug-associated toxicities in the entire cohort

Event	All cohort	CHIP		No CHIP		p value*
		Any grade	Grade 3 or 4	Any grade	Grade 3 or 4
Any adverse event, n (%)	60 (78.9)	34 (97)	30 (85.7)	33 (80)	15 (36.5)
Hematologic adverse event
Neutropenia	33 (43.4)	20 (57)	15 (42.8)	13 (31.7)	10 (24.3)	Any grade: 0.19
Thrombocytopenia	12 (15.7)	8 (22.8)	7 (20)	4 (0.9)	0	Grade 3 or 4 : 0.67
Anemia	10 (13.1)	9 (25.7)	7 (20)	1 (0.2)	0
Non-hematologic adverse event
GI	20 (26.3)	15 (42.8)	5 (14.2)	5 (12)	2 (4.8)	Any grade: 0.20
PN	15 (19.7)	12 (34.2)	9 (25.7)	3 (0.7)	0	Grade 3 or 4 : 0.19
Rash	8 (10.5)	6 (17.1)	6 (17.1)	2 (0.5)	1 (2.4)
Infections
Coronavirus disease 2019	20 (26.3)	16 (45.7)	5 (14.2)	4 (9)	0	Any grade: 0,36
Pneumonia	15 (19.7)	9 (25.7)	5 (14.2)	6 (14)	3 (7.3)	Grade 3 or 4 : 0.37
Cardio-vascular events
VTE	25 (32.8)	18 (51.4)	10 (28.5)	7 (17)	5 (12.1)	Any grade: 0,78
Ischemic events	5 (6.5)	4 (11.4)	1 (0.2)	1 (2.4)	0	Grade 3 or 4 : 0.62
Arrhythmias	8 (10.5)	5 (14.2)	2 (0.6)	3 (7.3)	2 (4.8)

PN: peripheral neuropathy
VTE: venous thromboembolism
GI: gastrointestinal toxicity *ChiSq test

Table 5

Frailty analysis of entire cohort according to Gordon Cook et al. Leukemia. 2020
	All cohort	CHIP	No CHIP	p value*
ECOG PS, n (%)				0.0005
0	8 (10.5)	0	8 (19.5)
1	37 (48.6)	14 (40)	23 (56)
≥ 2	31 (40.7)	21 (60)	10 (24.3)
Charlson Comorbidity Index, n				0.22
≤ 1	6 (7.8)	1 (2.8)	5 (12.1)
> 1	70 (92.1)	34 (97.1)	36 (87.8)
Age at diagnosis, n				1
≤ 75	54 (71)	25 (71.4)	29 (70.7)
76–80	9 (11.8)	4 (11.4)	5 (12.1)
> 80	13 (17.1)	6 (17.1)	7 (17)
Frailty definition				< 0.0001
Nonfrail	19 (25)	1 (2.8)	18 (44)
Frail	57 (75)	34 (97.1)	23 (56)
*Fisher Test

Platelets count positively correlates with VAF in MM patients

To analyze the specific impact of clonal hematopoiesis among our cohort of MM patients, correlogram and principal component analyses (PCA) were performed based on available clinical data. As shown in Fig. S1, CHIP and no-CHIP samples were projected onto the principal component (PC) space and two ellipses were calculated. Unfortunately, the first two PC1 to PC2, didn’t show a clear pattern thus suggesting that a clinical analysis-based approach fails to predict CHIP presence. CHIP refers to cancer-associated driver mutations present at mutational burden (VAF) ≥ 2% in subjects without hematologic abnormalities.⁵ Based on this assumption, we built a model to screen VAF and clinical data in our cohort. The chosen mathematical model turned out more effective than previous approach, by identifying platelet count as clinical indicator for greater VAF in MM patients regardless of specific carried-variants; (Fig. 7A) importantly, these results were also confirmed by focusing on the most frequent mutants DNMT3A and TET2. (Fig. 7B-C and S2). Overall these results led us to speculate platelets count as surrogate biomarker for higher VAF rate among CHIP carriers.

Recent studies have identified, in the blood of aging people, the presence of somatic mutations in hematopoietic cells, a condition referred as clonal hematopoiesis of unknown significance (CHIP).³⁶ By providing a fitness advantage to HSCs, CHIP is associated with a 0.5–1% risk of progression to a non-plasma-cell hematologic neoplasm such as MDS and AML.^5,7,14,37 Moreover, CHIP is also associated with aging-related conditions linked to aberrant inflammatory response such as atherosclerosis and cardiovascular diseases.¹⁸ In cancers, the prevalence of CHIP is higher compared to healthy population, especially for those patients exposed to cytotoxic chemotherapy or radiation.¹⁰ Recent studies suggest the presence of these myeloid clones also among MM patients with an impact on clinical outcomes;^{23–25,32,38,39} however, biological relevance of this abnormality still remains to be elucidated.

Here, we report the prevalence of CHIP in a multicenter cohort of NDMM patients at time of diagnosis and after 12–24 months of treatment, and describe the association of CHIP with clinical characteristics and outcomes of these patients. Overall we found somatic mutations in 46% of our population which is quite higher than expected and that reported earlier (21.6%).²⁵ These finding could be related to the sequencing used method, which achieves greater reading depth and therefore allows recording lower VAF (> 1%). Remarkably, consistent with prior studies, we found that CHIP status correlates with higher disease aggressiveness but differences were observed in term of aging, since previous studies have demonstrated CHIP prevalence with increasing age.^24,25 Although these findings appear in contrast, greater burden (VAF > 0.1) occurred among patients older than 70 years which is consistent with other studies⁴⁰ as well as for the most frequently mutated genes (DNMT3A followed by TET2). Moreover, we confirm association between poorer clinical outcomes, including progression-free and overall survivals which in turn reflects disease biologic status, as in part reported.^35,41,42

During recent years, novel drugs have significantly improved MM patient’s prognosis, but unfortunately continuous treatments have often limitations due to the development of serious and unexpected toxicities leading their discontinuation. Consistent with this observation, we analyzed, among CHIP carriers, those patients who experienced AEs during treatment: the event-free survival is greatly impaired by presence of CHIP with AEs of any grades, including infections, occurring more frequently in this group; similarly, recurrent infections were censured among CHIP carriers. Although the small sample size prevents any firm conclusion, we might speculate that a pro-inflammatory microenvironment may have impaired patient’s fitness leading to increased susceptibility to therapeutic-toxicity. Indeed, a detailed Bone Marrow environment focused on analysis revealed a consistent T cells impairment in CHIP carriers which, together with their higher frailty scores, explains the greater vulnerability to infections of this group. Moreover, while our data support CHIP screening at disease onset to better profile patient fitness, they also suggest its presence may influence T-cells based therapies, thus making CHIP evaluation a mandatory strategy to select the most appropriate therapeutic strategy for each MM patient.^43,19

Interestingly, our study includes also longitudinal data for a subset of patients: a slight increase in VAF was recorded for MDS/AML drivers (DNMT3A, TET2, ASXL1 and TP53) without appearance of additional mutations in other genes. This data demonstrates that micro-clones in the hematopoietic compartments of MM patients slightly increase in size but remain quite stable in terms of mutated genes type during disease evolution, thus suggesting that the mutational profile of myeloid cells may be largely unaffected by anti-MM therapies, including daratumumab-based regimens. On the other side, the presence of theses clones, as indicated by our data, correlates with a more aggressive disease and, worse outcome. As result, we speculate that the pro-inflammatory status, supported by these clones, facilitates tumor growth within BM microenvironment.

Recent studies suggest an impaired regenerative potential of hematopoietic stem cell grafts in autologous stem cell transplant recipients harboring CHIP. Namely, reduced stem cell yields and delayed platelet count recovery following ASCT is observed in presence of DNMT3A and PPM1D variants.³⁸ Consistent with these data, we found striking association between platelets count and CHIP burden: high platelets level indicates greater VAF in CHIP carriers, regardless of specific mutants. As reported, CHIP is associated with a pro-inflammatory status which lead to fourfold increase of cardiovascular illness incidence.⁴⁴ Biologic mechanisms of these dependencies are complex but, increased level of several cytokines (i.e. IL-6, IL-1 beta, IL-8 and NLRP3) are shared features. As result, inflammasome targeting agents (i.e. vitamin C and aspirin) are currently tested in ongoing clinical trials as preventive strategies to block CHIP-inflammation cascade and improve outcomes of individuals with CHIP. (ClinicalTrials.gov Identifier: NCT06097663, NCT03682029) Here, although the limited samples size of our cohort, we pinpoint relevance of platelets count, a well-known inflammatory indicator light, as promising biomarker of VAF among CHIP carriers. Importantly, by linking inflammatory-prone status and higher VAF rate, our study suggests the predictive role of platelets count to timely identify the greater tumor aggressiveness observed among CHIP carriers compared to other MM patients. However, the impaired fitness status observed among these patients, make us to speculate that less intensive therapies should be preferred in case of high platelets count to reduce severe adverse events occurrence, especially after high-dose regimens. However, larger prospective studies are needed to clarify the impact of platelet counts on those MM patients carrying CHIP and to support a cause-effect relationship with this association.

In conclusion, our study confirms that CHIP is frequently observed among MM patients where it correlates with a more aggressive disease and poorer clinical outcome as well. Interestingly, the presence of CHIP is associated with the early development of toxic events while its burden seems to be related to platelets count. Consistent with our results, we propose a specific management of CHIP-carrying MM patients, especially for what concern therapy-related toxicity.

ACKNOWLEDGMENTS

This work was supported in part by the Associazione Italiana per la Ricerca sul Cancro (AIRC, IG #23438, to M.C.), International Myeloma Society (IMS) and Paula and Rodger Riney Foundation Translational Research Award 2023.

AUTHORSHIP STATEMENT

E.G. and M.C. designed the research and wrote the manuscript; C.M., D.S., G.G., and I.T. performed samples and libraries preparation and collaborated to sequencing analyses; D.T. and F.L. performed the statistical and bioinformatics analyses; C.C., F.G., M.M., A.L, S.A., A.C and D.D. provided patient samples. F.DR., D.C., and R.M.L. revised the final version of manuscript.

DISCLOSURE OF CONFLICTS OF INTEREST

No conflicts of interest to disclose.

DATA AVAILABILITY

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Kumar, S. K. et al. Multiple myeloma. Nat. Rev. Dis. Primers. 3, 17046 (2017).
Rajkumar, S. V. Multiple myeloma: 2020 update on diagnosis, risk-stratification and management. Am. J. Hematol. 95, 548–567 (2020).
Marcon, C. et al. Experts’ consensus on the definition and management of high risk multiple myeloma. Front. Oncol. 12, 1096852 (2022).
Larocca, A. et al. Patient-centered practice in elderly myeloma patients: an overview and consensus from the European Myeloma Network (EMN). Leukemia. 32, 1697–1712 (2018).
Steensma, D. P. et al. Clonal hematopoiesis of indeterminate potential and its distinction from myelodysplastic syndromes. Blood. 126, 9–16 (2015).
Solís-Moruno, M., Batlle-Masó, L., Bonet, N., Aróstegui, J. I. & Casals, F. Somatic genetic variation in healthy tissue and non-cancer diseases. Eur. J. Hum. Genet. 31, 48–54 (2023).
Jaiswal, S. et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl. J. Med. 371, 2488–2498 (2014).
Bejar, R., CHIP & ICUS CCUS and other four-letter words. Leukemia. 31, 1869–1871 (2017).
Young, A. L., Challen, G. A., Birmann, B. M. & Druley, T. E. Clonal haematopoiesis harbouring AML-associated mutations is ubiquitous in healthy adults. Nat. Commun. 7, 12484 (2016).
Coombs, C. C. et al. Therapy-Related Clonal Hematopoiesis in Patients with Non-hematologic Cancers Is Common and Associated with Adverse Clinical Outcomes. Cell. Stem Cell. 21, 374–382e4 (2017).
Gibson, C. J. et al. Clonal Hematopoiesis Associated With Adverse Outcomes After Autologous Stem-Cell Transplantation for Lymphoma. J. Clin. Oncol. 35, 1598–1605 (2017).
Gillis, N. K. et al. Clonal haemopoiesis and therapy-related myeloid malignancies in elderly patients: a proof-of-concept, case-control study. Lancet Oncol. 18, 112–121 (2017).
Siegel, R., Naishadham, D. & Jemal, A. Cancer statistics, 2013. CA Cancer J. Clin. 63, 11–30 (2013).
Xie, M. et al. Age-related mutations associated with clonal hematopoietic expansion and malignancies. Nat. Med. 20, 1472–1478 (2014).
Dorsheimer, L. et al. Association of Mutations Contributing to Clonal Hematopoiesis With Prognosis in Chronic Ischemic Heart Failure. JAMA Cardiol. 4, 25–33 (2019).
Miller, P. G. et al. Association of clonal hematopoiesis with chronic obstructive pulmonary disease. Blood. 139, 357–368 (2022).
Agrawal, M. et al. TET2-mutant clonal hematopoiesis and risk of gout. Blood. 140, 1094–1103 (2022).
Jaiswal, S. et al. Clonal Hematopoiesis and Risk of Atherosclerotic Cardiovascular Disease. N Engl. J. Med. 377, 111–121 (2017).
Miller, P. G. et al. Clonal hematopoiesis in patients receiving chimeric antigen receptor T-cell therapy. Blood Adv. 5, 2982–2986 (2021).
Bejar, R. et al. Clinical effect of point mutations in myelodysplastic syndromes. N Engl. J. Med. 364, 2496–2506 (2011).
Cancer Genome Atlas Research Network et al. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl. J. Med. 368, 2059–2074 (2013).
Papaemmanuil, E. et al. Clinical and biological implications of driver mutations in myelodysplastic syndromes. Blood. 122, 3616–3627 (2013). quiz 3699.
Borsi, E. et al. Single-Cell DNA Sequencing Reveals an Evolutionary Pattern of CHIP in Transplant Eligible Multiple Myeloma Patients. Cells. 13. 10.3390/cells13080657 (2024).
Mouhieddine, T. H. et al. Clinical Outcomes and Evolution of Clonal Hematopoiesis in Patients with Newly Diagnosed Multiple Myeloma. Cancer Res. Commun. 3, 2560–2571 (2023).
Mouhieddine, T. H. et al. Clonal hematopoiesis is associated with adverse outcomes in multiple myeloma patients undergoing transplant. Nat. Commun. 11, 2996 (2020).
Norris, J. R. et al. Correlation of paramagnetic states and molecular structure in bacterial photosynthetic reaction centers: the symmetry of the primary electron donor in Rhodopseudomonas viridis and Rhodobacter sphaeroides R-26. Proc. Natl. Acad. Sci. U S A. 86, 4335–4339 (1989).
Becherini, P. et al. CD38-Induced Metabolic Dysfunction Primes Multiple Myeloma Cells for NAD+-Lowering Agents. Antioxidants (Basel) ; 12. doi: (2023). 10.3390/antiox12020494
Landrum, M. J. et al. ClinVar: improving access to variant interpretations and supporting evidence. Nucleic Acids Res. 46, D1062–D1067 (2018).
Rentzsch, P., Schubach, M., Shendure, J. & Kircher, M. CADD-Splice-improving genome-wide variant effect prediction using deep learning-derived splice scores. Genome Med. 13, 31 (2021).
Jaiswal, S., Natarajan, P. & Ebert, B. L. Clonal Hematopoiesis and Atherosclerosis. N Engl. J. Med. 377, 1401–1402 (2017).
DeZern, A. E., Malcovati, L. & Ebert, B. L. CHIP, CCUS, and Other Acronyms: Definition, Implications, and Impact on Practice. Am. Soc. Clin. Oncol. Educ. Book. 39, 400–410 (2019).
Rhee, J-W. et al. Clonal Hematopoiesis and Cardiovascular Disease in Patients With Multiple Myeloma Undergoing Hematopoietic Cell Transplant. JAMA Cardiol. 9, 16–24 (2024).
Testa, S. et al. Prevalence, mutational spectrum and clinical implications of clonal hematopoiesis of indeterminate potential in plasma cell dyscrasias. Semin Oncol. 49, 465–475 (2022).
Meier, J., Jensen, J. L., Dittus, C., Coombs, C. C. & Rubinstein, S. Game of clones: Diverse implications for clonal hematopoiesis in lymphoma and multiple myeloma. Blood Rev. 56, 100986 (2022).
Cook, G., Larocca, A., Facon, T., Zweegman, S. & Engelhardt, M. Defining the vulnerable patient with myeloma-a frailty position paper of the European Myeloma Network. Leukemia. 34, 2285–2294 (2020).
Jaiswal, S. & Ebert, B. L. Clonal hematopoiesis in human aging and disease. Science. 366. 10.1126/science.aan4673 (2019).
Genovese, G. et al. Clonal hematopoiesis and blood-cancer risk inferred from blood DNA sequence. N Engl. J. Med. 371, 2477–2487 (2014).
Stelmach, P. et al. Clonal hematopoiesis with DNMT3A and PPM1D mutations impairs regeneration in autologous stem cell transplant recipients. Haematologica. 108, 3308–3320 (2023).
Li, N. et al. Clonal haematopoiesis of indeterminate potential predicts delayed platelet engraftment after autologous stem cell transplantation for multiple myeloma. Br. J. Haematol. 201, 577–580 (2023).
Rodriguez, J. E., Micol, J. B. & Baldini, C. Exploring clonal hematopoiesis and its impact on aging, cancer, and patient care. Aging. 15, 14507–14508 (2023).
Zweegman, S. & Larocca, A. Frailty in multiple myeloma: the need for harmony to prevent doing harm. Lancet Haematol. 6, e117–e118 (2019).
Mian, H. et al. The prevalence and outcomes of frail older adults in clinical trials in multiple myeloma: A systematic review. Blood Cancer J. 13, 6 (2023).
Larocca, A., Cani, L., Bertuglia, G., Bruno, B. & Bringhen, S. New Strategies for the Treatment of Older Myeloma Patients. Cancers (Basel). 15. 10.3390/cancers15102693 (2023).
Libby, P. & Ebert, B. L. CHIP (Clonal Hematopoiesis of Indeterminate Potential): Potent and Newly Recognized Contributor to Cardiovascular Risk. Circulation. 138, 666–668 (2018).

Table 3 is available in the Supplementary Files section.

No competing interests reported.

Clonal Hematopoiesis Impacts Frailty of Newly Diagnosed Multiple Myeloma Patients: A Retrospective Multicentric Analysis

Status:

Version 1

Abstract

Figures

INTRODUCTION

PATIENTS AND METHODS

STUDY DESIGN

SAMPLE PREPARATION, SEQUENCING AND DATA ANALYSIS

MULTIPARAMETER FLOW CYTOMETRY (MFC) ANALYSIS

STATISTICAL ANALYSIS

RESULTS

Prevalence of Clonal Hematopoiesis among NDMM patients

CHIP is associated with greater aggressiveness and poorer clinical outcomes

Dynamic changes of Clonal Hematopoiesis during MM progression

CHIP presence enhances MM patients Frailty

Platelets count positively correlates with VAF in MM patients

DISCUSSION

Declarations

References

Table

Additional Declarations

Supplementary Files

Status:

Version 1