MYC protein is a high-risk factor in mantle cell lymphoma and identifies cases beyond morphology, proliferation and TP53/p53 – a Nordic Lymphoma Group study

Abstract

The transcription factor MYC is a well-described oncogene with an important role in lymphomagenesis, but its significance for clinical outcome in mantle cell lymphoma (MCL) remains to be determined. We performed an investigation of the expression of MYC protein in a cohort of 251 MCL patients complemented by analyses of structural aberrations and mRNA, in a sub-cohort of patients. Fourteen percent (n=35) of patients showed high MYC protein expression with >20% positive cells (MYC^high), among whom only one translocation was identified, and 86% (n=216) of patients showed low MYC protein expression. Low copy number gains of MYC were detected in ten patients, but with no correlation to MYC protein levels. However, MYC mRNA levels correlated significantly to MYC protein levels with a R² value of 0.76. Patients with a MYC^high tumor had both an independent inferior overall survival and an inferior progression-free survival (hazard ratio [HR]=2.03, 95% confidence interval [95% CI]: 1.2-3.4 and HR=2.2, 95% CI: 1.04-4.6, respectively) when adjusted for additional high-risk features. Patients with MYC^high tumors also tended to have additional high-risk features and to be older at diagnosis. A subgroup of 13 patients had concomitant MYC^high expression and TP53/p53 alterations and a substantially increased risk of progression (HR=16.9, 95% CI: 7.4-38.3) and death (HR=7.8, 95% CI: 4.4-14.1) with an average overall survival of only 0.9 years. In summary, we found that at diagnosis a subset of MCL patients (14%) overexpressed MYC protein, and had a poor prognosis but that MYC rearrangements were rare. Tumors with concurrent MYC overexpression and TP53/p53 alterations pinpointed MCL patients with a dismal prognosis with a median overall survival of less than 3 years. We propose that MYC needs to be assessed beyond the current high-risk factors in MCL in order to identify cases in need of alternative treatment.

Introduction

MYC is a pleiotropic transcription factor that regulates 10-15% of the genome and can simultaneously affect protein-coding genes and non-coding RNA products.^1,2 It is involved in a plethora of essential cellular mechanisms, such as cell growth, metabolism and protein synthesis, cell adhesion, apoptosis, cell cycle, and angiogenesis.^1-3 MYC is an important factor in B-cell proliferation known to be involved in lymphomagenesis,³ and its deregulation is frequently associated with worse outcome.^2-4

Translocations involving MYC and immunoglobulin genes are described in other B-cell lymphomas, particularly Burkitt and double-hit lymphomas, and are associated with aggressive behavior of the malignancy.² Overexpression of MYC without genetic rearrangements also has a negative impact on outcome in B-cell lymphomas in general.⁵ In mantle cell lymphoma (MCL), mutations of MYC have been shown to occur in around 20% of cases,⁶ and MYC-regulated pathways are often affected by genetic alterations in subsets of disease.⁷ Cytogenetic investigations of MYC were recently recommended as part of clinical routine but are not yet clinically implemented in most hospitals.⁸ The frequency of translocations involving MYC is reportedly low^9-11 and most reports have been case studies describing translocations that juxtapose MYC and CCND1.^12-15 In general, consensus on the degree of MYC deregulation in MCL is lacking. Here, we aim to describe the frequency of protein expression, mRNA, translocations, and amplifications of MYC in primary MCL and relate the findings to clinicopathological parameters and outcome.

MCL is a disease with a heterogeneous clinical behavior characterized by the chromosomal t(11;14)(q13;q32) that juxtaposes CCND1 to immunoglobulin genes, leading to constitutive cyclin D1 overexpression.¹⁶ An established prognostic tool, the MCL International Prognostic Index (MIPI), integrates information on age, performance status, lactate dehydrogenase levels and white blood cell count and stratifies patients into high, intermediate and low risk.¹⁷ Additional biological risk factors include TP53 mutations and/ or p53 overexpression, high proliferation, and non-classic morphology.¹⁶ Despite recent improvements in treatment,¹⁸ MCL patients often have a poor prognosis and frequently relapse.¹⁹ Thus, identification and improved understanding of alterations in MCL lymphomagenesis, beyond the already established risk factors, are critical in order to be able to individualize therapeutic decision-making.

To date, a limited number of studies have focused on MYC in MCL. In 2017, Hu et al. showed that MYC rearrangements were present in less than 1.0% of MCL cases at diagnosis, with these cases having a median overall survival (OS) of 31.3 months.²⁰ When compared with other subtypes of lymphoma, MCL tumors seem to have a lower frequency of structural alterations involving MYC.²¹ Nonetheless, in an evaluation of 88 patients, Wang et al. described 27 with tumors with MYC rearrangements and 21 with extra copies of MYC. Both subgroups of patients showed a lower OS when compared to cases with no MYC aberrations.²² However, the study by Wang et al., and other investigations reporting MYC translocations, were selected cohorts enriched for structural alterations involving MYC.^9,20,23 An extensive study evaluating 1,214 lymphomas, including 138 cases of MCL, did not find any MYC rearrangements in MCL cases and only 2% had MYC protein overexpression >26%.¹¹

Overall, MYC aberrations in MCL have been associated with a worse prognosis,^10,23,24 non-classic morphology^20,21,23,24 and, reportedly, enrichment for p53 overexpression among MYC-overexpressing tumors,^24-26 but their additive role in MCL prognosis and clinical characteristics has not been determined. A study in diffuse large B-cell lymphoma showed that tumors with dual MYC/TP53 alterations had distinct clinicopathological characteristics with worse survival compared to wild-type cases.²⁷

In the current study, we investigated the frequency of each molecular layer of MYC deregulation in primary diagnostic samples from a cohort of MCL patients and identified its association with both clinical and molecular high-risk factors.

Methods

Patients’ material

Two hundred and fifty-two MCL patients were included in this study, and 154 patients were part of a population-based cohort that comprised patients registered in the Swedish Lymphoma Register (SLR) and diagnosed in Uppsala and Southern Sweden between 2000-2017. The additional 98 samples were derived from patients enrolled in the Nordic Lymphoma Group clinical trials MCL2 and MCL3 (N-MCL2/3) (Online Supplementary Figure S1). The N-MCL2 trial was registered with ISRCTN.com ID ISRCTN87866680; the N-MCL3 trial was registered with ClinicalTrials.gov ID NCT00514475. Further details are provided in the Online Supplementary Materials and Methods.

This study was approved by the Ethical Regional Committee in Lund (Dnr 2011/593) for part of the SLR cohort (BLISS) and by the Ethical Regional Committee in Uppsala for the N-MCL2/3 samples (Dnr 2009/428) and for part of the SLR cohort (U-CAN) (Dnr 2014/233).

Immunohistochemistry

The patients’ tissue microarrays were stained with anti-MYC antibody (clone Y69 1:50, Abcam; Cambridge, UK), as used in the clinical setting and in previous publications.¹⁰,²⁴ Tissue samples were considered to overexpress MYC when the percentage of cells with a dark brown nucleus was ≥20%, in agreement with published data.²⁴ Details about the immunohistochemistry investigaions are provided in the Online Supplementary Materials and Methods.

Fluorescent in-situ hybridization

Fluorescent in-situ hybridization (FISH) was performed on 4 μm tissue sections using split-signal DNA probes for MYC with Vysis MYC Break Apart FISH Probe (Abbott Laboratories; Green Oaks, IL, USA) according to instructions from the manufacturer. An Olympus BX-51 microscope (Prior Lumen200 light source) and GenASIs Capture and Analysis Platform software (Applied Spectral Imaging; Carlsbad, CA, USA) were used to capture digital images of tumor areas. Cases without representative tumor material or with no representative signals were excluded. Positive FISH results on tissue microarrays were validated by evaluating whole-tissue sections.

mRNA in situ hybridization

MYC mRNA was evaluated in 85 fresh-frozen, paraffin-embedded samples from the SLR cohort with the RNAscope® assay (Advanced Cell Diagnostics; Newark, CA, USA) following the manufacturer’s instructions. The H-score was calculated for each sample, and was defined as the dynamic range of MYC expression based on the quantification of the probe signal on a cell-by-cell level. The workflow is described in detail in the Online Supplementary Materials and Methods.

Multiplexed immunofluorescence staining

Tissue microarray slides were stained with anti-CD20 (11.9 µg/mL, clone IGEL/773, Novus Biologicals, Littleton, CO, USA), anti-CD3 (2 µg/mL, clone UM500048CF, OriGene; Rockville, MD, USA) conjugated with AlexaFluor 532 antibody labeling kit (Thermo Fisher Scientific, Waltham, MA, USA), anti-CD163 (1.25 µg/mL, clone EPR14643-36, Abcam; Cambridge, UK) conjugated with AlexaFluor 647 antibody labeling kit (Thermo Fisher Scientific), and Syto13 (500 nM, Nanostring, Seattle, WA, USA).

A software for deep learning artificial intelligence, Aiforia Create Version 5.3 (Aiforia Technologies Plc, Helsinki, Finland), was used for image analyses as further described in the Online Supplementary Materials and Methods.

Statistical analysis

A χ² test, t test or Wilcoxon signed-rank test was used to evaluate differences between groups. The Pearson correlation coefficient was applied to test correlations between continuous variables. The outcome variables considered in the study were OS and PFS. Maximally selected rank statistics (Max Rank) in R²⁸ was used to determine a cutoff for MYC. Differences were considered statistically significant when the P value was <0.05. A detailed description of the statistical analysis is provided in the Online Supplementary Materials and Methods.

Results

Patients’ clinicopathological characteristics

A total of 252 patients were included in this study (Table 1), with 98 patients belonging to the N-MCL2/3 clinical trials and 154 patients being part of the SLR cohort. Male patients were predominant (75%) and the median age at diagnosis was 63 years. Patients were evenly distributed among the different MIPI risk groups, but this information was only available for 196 of the 252 patients. Thirty-one percent (76/242) of the patients had highly proliferative tumors (Ki67 >30%) and 16% (38/242) had non-classic morphology. A subgroup of patients had tumors with high expression of p53 (12%, 29/240) and/or TP53 mutations (20%, 24/119). Thus, 35/242 cases had either high p53 protein expression or a TP53 mutation. The discrepancy in frequency between the p53 and TP53 evaluation is mostly due to the lower number of samples sequenced, restricted by the availability of high-quality material. Overall, patients had a median OS of 6.2 years and a median PFS of 4.8 years. Of note, OS information was available for all patients included, whereas PFS was calculated based on 200 patients. The patients’ clinicopathological characteristics, divided by cohort, can be found in Online Supplementary Table S1.

Table 1.

Clinicopathological characteristics of the patients included in this study.

MYC overexpression

Based on previous studies in which MYC expression was assessed in a cohort of 65 cases of MCL,²⁴ a cutoff of 20% was used to define patients with MYC protein overexpression (MYC^high). Immunohistochemistry was used to determine MYC protein expression. The mean expression was 13.1% (range, 0.14-82.9%) and the median expression was 8.7%. Max Rank statistics showed that OS outcome differences were maximized when groups were dichotomized with a cutoff at 21.4%, supporting the applicability of the 20% cutoff used to define MYC overexpression.²⁴ Using the 20% cutoff, MYC was overexpressed in 14% of all tumor samples studied (35/252) (Table 1). Online Supplementary Figure S2 shows representative immunohistochemistry staining for MYC.

MYC mRNA and immunohistochemistry results are concordant

To evaluate MYC mRNA expression levels, we performed RNAscope^®; representative images of the staining are shown in Online Supplementary Figure S3. An H-score for MYC mRNA expression was calculated, which showed a strong correlation (Pearson) to the frequency of positive cells detected by immunohistochemistry (Online Supplementary Figures S4 and S5) with a R² value of 0.76 (P<0.001), in the same range as previously reported for MCL.¹⁰ Information about the cohort used for RNAscope^® can be found in Online Supplementary Table S2.

MYC^high is associated with poor outcome

Outcome in patients with MYC^high tumors was inferior compared to that of patients with MYC^low tumors (Figure 1). Patients with MYC^high MCL had a median OS of 2.2 years and PFS of 1.8 years, whereas patients with MYC^low tumors had a median OS of 7.3 years and PFS of 5.2 years. In concordance, patients with MYC^high tumors also had a significantly higher risk of death (hazard ratio [HR]=2.34, 95% confidence interval [955 CI]: 1.55-3.57) and disease progression (HR=1.73, 95% CI: 1.05-2.86) compared to patients with MYC^low tumors. The frequency of MYC-positive cells was also significantly associated with poor outcome as a continuous variable, for both OS and PFS. However, per increased percentage of MYC-positive cells, the risk of death or progression only increased by 3% and 4%, respectively (Table 2). Likewise, for MYC mRNA (as a continuous value), a higher H-score was associated with inferior survival and shorter progression-free survival, albeit with a low hazard ratio per H-score unit increase (Table 2). MYC protein measurements violated the proportional hazards assumption, showing a greater impact on short-term survival (Online Supplementary Figure S6). Of the cases classified as MYC^high, 57% died and 62% progressed within the first 3 years (Figure 1). In a follow-up of 3 years, the proportional hazards assumption for PFS was not violated and a negative impact on survival could be observed (Online Supplementary Table S3). Similar trends were seen when evaluating the effect of MYC^high and MYC frequency as a continuous value on outcome for the different cohorts separately (Online Supplementary Figure S7, Online Supplementary Table S4). However, in the SLR cohort, MYC^high was not statistically prognostic for PFS, probably because of the heterogenous treatment protocols used in this population-based cohort. In the N-MCL2/3 cohort, only six patients were classified as MYC^high and the group reached statistical significance as predictor for PFS, but not OS (Online Supplementary Table S4).

Figure 1.

Patients with tumors with MYC^high protein overexpression have a worse outcome than patients with MYC^low tumors. (A, B) Prognostic impact of MYC protein expression on overall survival (A) and progression-free survival (B). Samples (N=252) were categorized based on total percentage of MYC-expressing cells, with a cutoff of 20%. Kaplan-Meier estimates were calculated and are shown. Log-rank statistics were used to evaluate the statistical significance. PFS: progression-free survival.

To understand if there were consistent differences among MYC^high cases depending on whether they had an OS shorter or longer than 3 years, we compared the main clinicopathological parameters of these two groups of patients (Online Supplementary Table S5). The patients with an OS less than 3 years had additional high-risk factors, being in a high-risk MIPI group and the majority also with high proliferation and/ or TP53/p53 aberrations. Surprisingly, the male/female ratio showed a major difference, with only one out of nine female MYC^high patients surviving for more than 3 years.

MYC^high tumors remained associated with OS (HR=2.03, 95% CI: 1.22-3.40) and PFS (HR=2.20, 95% CI: 1.04-4.64) when adjusting for gender, age and established high-risk factors (Table 3). MYC^high was not significant when adjusting for MIPI group (Table 3). Of note, evaluation of MYC protein expression on prognosis in MCL remained significant only when considering patients with MIPI information (Online Supplementary Table S6). MYC as a continuous variable was not independent of high-risk factors (data not shown).

MYC^high is associated with high-risk factors

Patients with MYC^high tumors were enriched in other high-risk factors such as age, Ki-67 proliferation index, non-classic morphology, MIPI group and TP53/p53⁺ aberrations (Figure 2A, Table 4). Patients with MYC^high tumors were on average older at diagnosis (Figure 2B), with a median age at diagnosis of 70.7 years versus 63.6 years for patients with MYC^low tumors. MYC protein overexpression (MYC^high) was found to be associated with the presence of TP53 mutations and/or p53 overexpression (hereon referred to as TP53/p53⁺ tumors, n=35). Dual alterations of MYC and TP53/p53⁺ (hereon referred to as MYC^high TP53/p53⁺) were detected in 13 out of 250 patients (Table 4). Tumors with alterations in TP53 had a higher median expression of MYC-positive cells compared to wild-type tumors (P<0.001) (Figure 2C) and MYC^high TP53/ p53⁺ cases were more likely to have high proliferation, non-classic morphology and be in a high-risk MIPI group, similar to the MYC^high group (Table 4).

MYC^high TP53/p53⁺ tumors have a dismal prognosis

A total of 13 patients had tumors classified as MYC^high TP53/ p53⁺. These patients had a very short median survival (0.9 years) and a shorter median PFS (0.5 years) compared to patients who did not show concomitant alterations in these molecules. The majority died within 3 years of diagnosis (Figure 3A). This was significantly lower compared to either one of the individual high-risk groups (Figure 3B). All patients with PFS information available with MYC^high TP53/p53⁺ tumors progressed within 2 years (Figure 3C) and had a shorter time to progression than either of the patient groups presenting with only one of the individual risk factors (Figure 3D). Online Supplementary Figure S8 depicts the process for classification of tumors based on MYC and TP53/p53 status. Double aberrations conferred an increased risk of death (HR=7.83, 95% CI: 4.35-14.09) and disease progression (HR=16.87, 95% CI: 7.43-38.31) (Table 2). MYC^high TP53/p53⁺ aberrations remained prognostic, for both OS and PFS, when adjusting for additional high-risk factors (Table 5). Of note, prognostic analysis considering only TP53 mutation and MYC^high tumors remained significant for both OS and PFS (Online Supplementary Figure S9).

MYC protein expression is not correlated with genomic aberrations

To explore the hypothesis that MYC protein overexpression can be associated with genomic rearrangements of MYC, a total of 85 cases mounted in tissue microarrays were evaluated by FISH with a MYC break-a-part probe. Of those 85 patients, 70 (82%) showed no signs of genomic alterations of MYC. Among the remaining 15 cases, three showed copy gains in less than 20% of the evaluated cells, 11 showed copy gains in more than 20% of the evaluated cells and one case showed the presence of a translocation (Online Supplementary Figure S10).

To validate the FISH findings, whole tissue sections were used. Diagnostic tissue blocks from 14 out of the 15 patients were available for further evaluation (Online Supplementary Table S7). Results were consistent between tissue microarrays and full tissue sections. No correlation between copy number gains of MYC and protein overexpression was observed, as most of tumors with MYC copy number gains had a low frequency of cells expressing MYC protein (Online Supplementary Table S7). The presence of copy number gains was not associated with outcome (data not shown). The single MYC-translocated MCL case had an OS of over 6 years, no other high-risk factors but a classic morphology, low Ki-67 expression, and was TP53/p53 wild-type.

Table 2.

Univariable Cox proportional hazards models results.

Table 3.

MYC protein multivariable Cox proportional hazards models.

Table 4.

Differences in clinicopathological characteristics with a focus on established high-risk factors comparing MYC^high tumors with MYC^low tumors and comparing MYC^highTP53/p53⁺ tumors to all others.

In line with the lack of genomic alterations, MYC showed genetic mutations in only one patient, determined by targeted sequencing of part of the SLR cohort as previously published.²⁹ This MYC mutated case only had 4% of cells expressing MYC protein.

MYC^high tumors showed an increase in M2-like macrophage infiltration

In a recent study, we showed that the frequencies of T cells and M2 macrophages are prognostic in MCL³⁰ and hypothesized that the composition of the microenvironment would be different in MYC^high and MYC^low cases. T-cell and M2 macrophage frequencies were determined based on multiplexed immunofluorescence staining using machine learning (Aiforia software). The multiplexed immunofluorescence staining included stains for CD3, CD163 and CD20. For a total of 117 patients, information was available for both multiplexed immunofluorescence and MYC status. T-cell infiltration, marked as CD3 positivity, ranged from 1.48% to 52.58% of total cells. The presence of CD163⁺ cells, as a surrogate marker for M2 macrophages, varied between 0.02% and 23.09% of total cells. The abundance of T cells and M2-like macrophages and the density of tumor cells in the tumor regions in MYC^high and MYC^low cases were compared to assess the association with MYC status. MYC^high cases had a higher infiltration of CD163⁺ macrophages, whereas no difference was seen for T-cell infiltration or density of tumor cells (Online Supplementary Figure S11).

Discussion

The impact of MYC deregulation on outcome in MCL is consistently reported in the literature,^10,22,23 but most studies include limited or selected cohorts. The present study of 252 MCL patients is, to the best of our knowledge, the largest study so far to evaluate clinical impact by exploring MYC protein expression and its association with clinicopathological features including established risk factors. The study is focused on the impact of MYC protein, and the association with gene amplifications, and rearrangements, and transcriptional activity.

Figure 2.

High-risk mantle cell lymphoma is enriched for MYC^high tumors. (A) Heatmap representing the distribution of patients with high and low MYC protein expression, TP53 mutation/p53 expression, morphology variant, proliferation (Ki-67) and age at diagnosis. (B) Association between age at diagnosis and MYC^high tumors. (C) Association between total percentage of MYC-positive cells and TP53/p53 status. Wilcoxon and t-test P values were used to evaluate statistical significance, and P values are shown. WT: wild-type; MIPI: Mantle cell lymphoma International Prognostic Index.

Figure 3.

Patients with MYC^highTP53/p53⁺ form a subgroup with adverse prognosis. (A) Prognostic impact of MYC^highTP53/p53⁺ versus all other states on overall survival. (B) Prognostic impact on overall survival for patients with tumors that were classified only as MYC^high, only TP53/p53⁺, both MYC^highTP53/p53⁺ or wild-type for both markers. (C) Prognostic impact of MYC^highTP53/p53⁺ versus all other patients on progression-free survival. (D) Prognostic impact on progression-free survival for patients with tumors that were classified only as MYC^high, only TP53/p53⁺, both MYC^highTP53/p53⁺ or wild-type for both markers. Kaplan-Meier estimates were calculated and are shown. Log-rank statistics were used to evaluate the statistical significance of differences. WT: wild-type; PFS: progression-free survival.

Table 5.

MYC^high TP53/p53⁺ multivariable Cox proportional hazards models.

Current literature on MYC in MCL has not reached a consensus on a cutoff for the definition of MYC^high.^10,25,31,32 Previous studies used 10%, 20% or 26%,^10,11,24 which hampers direct comparison of the impact of dichotomized groups. We applied a 20% cutoff based on the findings of a previous study on 65 MCL patients by Choe et al.²⁴ To validate the applicability of the applied cutoff in our cohort, we further used Max Rank statistics that identified 21% as the cutoff maximizing the difference in OS in the present cohort.

Most MCL patients had tumors with a low frequency of MYC-positive cells, but 14% (n=35) of the tumors had >20% positive cells and were defined as MYC^high. The patients with MYC^high tumors had a median OS of 2.2 years and PFS of 1.8 years. These were significantly shorter than the OS of 7.3 years and the PFS of 5.2 years for patients with MYC^low tumors. The impact on outcome is similar to that previously reported by Oberley et al.¹⁰ Both elevated mRNA and protein MYC levels were associated with poor prognosis. Of interest, MYC levels violated the proportional hazards assumption, indicating that the prognostic effect of MYC protein is time-dependent, with a mainly negative impact in the first 3 years after diagnosis. To understand whether there were other high-risk features that separated MYC^high patients with OS shorter or longer than 3 years, the clinicopathological characteristics were compared and confirmed that additional high-risk factors such as being in a high-risk MIPI group, having TP53/p53 aberrations and high proliferation were more common in MYC^high patients with an OS shorter than 3 years. Of note, women with MYC^high tumors seemed to do even more poorly than men, with only one out of nine women having an OS longer than 3 years. Although numbers are small, this indicates that there might be a sex-related difference in the adverse effects mediated by MYC.

Comparison of MYC^high and MYC^low tumors with clinicopathological parameters showed that there is a positive correlation between high age and MYC overexpression in MCL. This has not been reported before. In the study by Aukema et al.²⁶ most patients with high MYC expression were older than 65 years, although this was not specifically mentioned. High expression of MYC is also associated with older age in patients with anaplastic lymphoma kinase-positive large cell lymphomas³³ and diffuse large B-cell lymphoma.³⁴

Besides age, patients with MYC^high tumors were also enriched for other high-risk features such as high-risk MIPI group, non-classic morphology, and high proliferation. However, the negative impact of MYC on outcome was independent of these high-risk factors, with the exception of MIPI group, emphasizing the additive effect between molecular factors, which was explored further. The impact of TP53 mutations and/or p53 overexpression in MCL has been widely documented by us and others^35,36 and assessment of TP53 status should be performed in routine clinical diagnostics.¹⁶ Wild-type p53-mediated apoptosis can be induced by MYC,³⁷ and as MYC overexpression in cancer is believed to lead to deregulation of its physiological targets,³⁸ altered TP53/p53 could synergize with MYC and lead to a more aggressive variant of the disease. Here we show that simultaneous alterations in MYC and TP53/p53 (MYC^highTP53/p53⁺) did indeed have an additive negative prognostic effect compared to either of the aberrations alone, being associated with a median OS and PFS of only 0.9 and 0.5 years, respectively. The MYC^highTP53/p53⁺ subgroup of patients had unfavorable clinical characteristics, with most having highly proliferative tumors, with non-classic morphology and being in a high-risk MIPI group. The presence of tumors with both aberrations had been noted in other studies,^23,26,39,40 but this is the first time that an association with prognostic and clinicopathological parameters in MCL is reported. The negative effect of MYC overexpression combined with TP53 mutations, excluding patients with wild-type TP53 but high levels of p53 protein, was also confirmed. The mechanisms behind MYC and TP53/p53 crosstalk remain to be explored, but these aberrations are known to influence each other by involvement of proteins, such as BMI-1, ARF, and microRNA, including microRNA34a.⁴¹ Den et al. studied the synergetic effect of MYC and TP53/p53 abnormalities on outcome in patients with diffuse large B-cell lymphoma, similarly to this study on MCL.

In our cohort we identified only one case with MYC rearrangements, but a few cases (1.3% of the 85 evaluated cases) showed copy gains of MYC. These results are in line with prior studies in which MYC rearrangements were rare.^10,20,42 When genetic alterations have been observed, they have mainly been amplifications/copy gains rather than chromosomal translocations of MYC.³¹ A high frequency of MYC abnormalities has been found in only one selected cohort.²³ Similarly, concurrent translocations of MYC and CCND1 have been reported only as single cases, except in a selected cohort with overrepresentation of leukemic MCL in which 5% of MCL tumors had these dual aberrations.⁴³ Thus, we can conclude that translocations or other rearrangements involving MYC are rare in MCL at diagnosis, corresponding to less than 2% of all cases in most studies. However, in the present study MYC protein overexpression identified 14% of MCL patients with a poor prognosis and added information on high risk beyond TP53/p53, morphology and proliferation.

In other B-cell lymphomas, such as Burkitt lymphoma and diffuse large B-cell lymphoma, structural alterations involving MYC are a predominant mechanism leading to MYC overexpression.²⁵ In MCL, we found that MYC copy number gains were not correlated with outcome or MYC mRNA or protein overexpression. The expression seems to be driven by transcriptional dysregulation and mRNA and protein expression were highly correlated with similar effects on outcome. Investigations of MYC-driven lymphomagenesis in MCL support this notion, by showing that miRNA, such as miR33b, miR96, and miR503, are pivotal in the regulation of MYC⁴⁴ and that histone deacetylation is involved in the repression of transcription mediated by MYC.⁴⁵ In addition, MALT1 has also been proposed as an alternative mechanism for MYC protein stabilization in MCL.⁴⁶ Studies in other lymphomas show that hotspot mutations in regions that can affect protein stability are selected during lymphomagenesis and are associated with a negative impact on outcome.⁴⁷ It has also been shown by Nadeau et al. that MYC and TP53 are the only genes whose alterations have an impact on outcome beyond that of the total genomic complexity.⁴⁸ However, mutations of MYC do not seem to drive relapse in MCL as MYC has been shown to be less mutated in relapsed cases compared to cases at diagnosis.⁶ In other cancers, MYC has a role in shaping the tumor immune microenvironment, through several mechanisms. Indeed, MYC is capable of immune checkpoint regulation, like PD-L1 and MHC class I and II molecules, and promotes cytokine secretion, leading, as an example, to re-programing M1- to M2-like macrophages.^49,50 We hypothesized that MYC^high tumors would have an altered immune microenvironment. Image analyses showed no differences in T-cell infiltration but MYC^high tumors were associated with increased infiltration of CD163⁺ cells. This suggests that M2 macrophages may contribute to an adverse outcome in such tumors and that MYC is associated with both intrinsic and extrinsic high-risk features.

Previous studies in cancer have shown the potential of MYC inhibition to promote antitumor effects. Nonetheless, due to intrinsically disordered domains and lack of enzymatic sites, MYC has been considered undruggable.⁵¹ Several approaches have been proposed to inhibit MYC both directly and indirectly at all its levels of regulation. Recently, the first direct MYC inhibitor, the blocking peptide OMO-103 (Peptomyc), reached clinical phase studies. OMO-103 has been shown to alter the tumor microenvironment, potentiating an antitumor immune response,⁵² providing hope for future successful clinical use of this agent. MYC has been shown to have a role in ibrutinib resistance.^53,54 Thus, also in the ibrutinib era in MCL, we expect that MYC will remain a high-risk marker and that MYC-targeting therapies may play an important role in both the diagnostic and relapsed setting. The current study included 252 MCL patients evaluated at diagnosis; their treatment during the follow-up was not homogenous. As both MYC and p53 aberrations affect a limited group of patients, results need to be validated in independent cohorts of patients, ideally under the same treatment protocol. The analyses are further limited by the fact that no mutational analyses were performed, so no correlation between different mutational sites that may affect protein stability and protein expression could be identified.

In summary, MYC protein is a high-risk marker and, in this study, was overexpressed in a significant subgroup of cases of MCL (14%). Overexpression of MYC (>20% expression) adds prognostic information beyond that of established molecular risk factors, such as TP53/p53, morphology and proliferation, for risk stratification of MCL patients. MCL patients carrying tumors with both MYC^high and TP53/p53 aberrations constitute a subgroup with a dismal prognosis, indicating additive negative effects. Previous efforts at risk stratification have included the presence of upregulation of MYC at the bulk mRNA level together with other markers in a five-gene signature that predicts survival in MCL.⁵⁵ However, we propose that MYC may be assessed through routine immunohistochemistry, using a cutoff at 20% to separate high from low expression, together with routine assessment of TP53/p53 status, proliferation and MIPI group.

Funding Statement

Funding: This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Marie Skłodowska-Curie grant agreement N 754299, Cancerfonden (2016/465, 19 0309Pj and 21 1561 Pj), Mats Paulssons Stiftelse för forskning, innovation och samhällsbyggande, Stiftelsen Stefan Paulssons cancerfond, and CREATE Health. All financial support was granted to SE.

Data-sharing statement

Original data and protocols are available upon request.