Differential T-cell and monocyte responses in hepatocellular carcinoma treated with regorafenib plus nivolumab: an integrated clinical and biomarker analysis of the phase 2 RENOBATE trial

doi:10.21203/rs.3.rs-3113386/v1

Download PDF

Article

Differential T-cell and monocyte responses in hepatocellular carcinoma treated with regorafenib plus nivolumab: an integrated clinical and biomarker analysis of the phase 2 RENOBATE trial

https://doi.org/10.21203/rs.3.rs-3113386/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 19 Feb, 2024

Read the published version in Nature Medicine →

Version 1

posted

You are reading this latest preprint version

In this phase 2 REBNOBATE trial, we evaluated regorafenib-nivolumab as front-line treatment for unresectable hepatocellular carcinoma (uHCC). Patients (n = 42) received nivolumab 480 mg every 4 weeks, and regorafenib 80 mg daily (3-weeks on/1-week off schedule). Single-cell RNA sequencing was performed using peripheral blood mononuclear cells from early progressors (progressively increased tumor burden) and long-term responders (response/stable disease for ≥ 10 months). The overall response rate in the study was 31.0%, with a median progression-free survival of 7.4 months and a 1-year overall survival rate of 80.5%. Regorafenib-nivolumab was well-tolerated (no new safety signal). Long-term responders exhibited T-cell receptor repertoire diversification; enrichment of genes representing immunotherapy-responsiveness and cytotoxicity in MKI67⁺ proliferating CD8⁺ T cells; and interaction between MKI67⁺ proliferating CD8⁺ T cells and classical monocytes through IFN-γ pathways with a higher probability of M1-directed polarization of monocytes. Classical monocytes from early progressors exhibited upregulation of TMEM176A/B coupled with an ineffective inflammasome response. Regorafenib-nivolumab is feasible as front-line treatment for uHCC. Our findings may support the development of biomarkers, or novel immunotherapies to overcome resistance in uHCC.

Biological sciences/Cancer/Gastrointestinal cancer/Liver cancer

Biological sciences/Cancer/Tumour immunology

Hepatocellular carcinoma

regorafenib-nivolumab

pharmacologic biomarker

classical monocytes

proliferating CD8+ T cells

and immune evasion

In global phase 3 trials of first-line¹ and second-line² settings, anti-PD-1 antibody monotherapy has failed to yield overall survival (OS) benefit in patients with unresectable hepatocellular carcinoma (HCC). However, phase 3 IMBrave 150 study³ and HIMALAYA study⁴ demonstrated the OS benefit of first-line immune checkpoint inhibitor (ICI)-based combination regimens compared to sorafenib, which resulted in their approval. In a recent first-line phase 3 trial, a combination of the VEGFR2-targeted multi-kinase inhibitor (MKI) rivoceranib, plus the anti-PD-1 antibody camrelizumab, yielded significantly improved survival outcomes compared to sorafenib⁵. However, other anti-PD-1/PD-L1 plus MKI combination regimens, such as cabozantinib-atezolizumab and lenvatinib-pembrolizumab, have failed to show OS benefits compared to MKI monotherapy (sorafenib and lenvatinib, respectively) in first-line phase 3 trials for unresectable HCC^6,7. The reasons for these discrepancies are not yet clear, but the results suggest that clinical efficacy may partly depend on differences in the detailed molecular and/or immunological mechanisms of the combination partners, especially in the context of ICI-based combination treatment.

Regorafenib is a multi-kinase inhibitor with anti-angiogenic activity that inhibits various targets, including VEGFR 1–3, KIT, RET, RAF-1, BRAF, PDGFR, FGFR, and CSF-1R⁸. It has been shown to improve survival outcomes in patients with unresectable HCC, after progression with sorafenib treatment^9,10. Among its various targets, the ability of regorafenib to modulate the VEGFR and CSF1R pathways highlights its potential to reverse the immunosuppressive gradients of myeloid cells, and potentiate anti-tumor immune responses. In particular, inhibition of the CSF1R pathway could polarize myeloid cells towards exerting anti-tumor responses^11–13. Since myeloid cells are critically implicated in the immune evasion process in HCC^14–16 and in the ICI response¹⁷, there is a strong rationale for combining regorafenib and ICI in patients with HCC. Moreover, the regorafenib-nivolumab combination has shown promising efficacy and safety outcomes in a phase 1b trial for patients with gastric and colorectal cancer¹⁸. In the CheckMate459 study, nivolumab monotherapy was associated with a trend toward improved survival outcomes compared to sorafenib, but failed to meet the pre-defined statistical significance for improving survival in patients with unresectable HCC¹. The anti-tumor activity and potential immunomodulatory effects of regorafenib suggest that the combination of regorafenib and nivolumab may have clinically meaningful benefits in HCC patients.

Although some biomarkers are reportedly associated with clinical outcomes of HCC patients treated with ICI^19–21, none has yet been firmly established as predictive for ICI response. This suggests a need to further investigate the detailed mechanisms of response or resistance to ICI-based treatments.

The RENOBATE is a multicenter phase 2 study evaluating the efficacy and safety of the first-line regorafenib-nivolumab combination in patients with unresectable HCC. Herein, we present the results in terms of clinical outcomes and a comprehensive biomarker study, using circulating tumor DNA (ctDNA), single-cell RNA sequencing (scRNA-seq), single-cell T-cell repertoire (TCR) and multicolor flowcytometry using serially collected peripheral blood mononuclear cells (PBMCs) from the patients.

Clinical outcomes of regorafenib-nivolumab

A total of 42 patients were enrolled between July 2020 and January 2021 (Extended Data Fig. 1). The median age was 61 years (range, 40–79 years), and 31 patients (73.8%) were male (Supplementary Table 1). Most patients had BCLC C stage (n = 38, 90.5%), and hepatitis B virus infection as an etiology of HCC (n = 30, 71.4%). The primary endpoint of the clinical trial was objective response rate (ORR) assessed by Response Evaluation Criteria in Solid Tumors (RECIST) v1.1. ORR per RECIST v1.1 was 31.0%, meeting the primary endpoint (ORR of ≥ 25%) : complete response (CR) in 1 patient (2.4%) and partial response (PR) in 12 patients (28.6%) (Fig. 1a,b). ORR per mRECIST was 33.3%, with CR in 2 patients (4.8%) and PR in 12 patients (28.6%). With a median follow-up duration of 11.1 months (95% CI, 6.11–14.0 months), the median PFS was 7.38 months (95% CI, 4.12–13.0 months) and the 1-year PFS rate was 37.8% (Fig. 1c). The 1-year OS rate was 80.5% (95% CI, 63.0–90.3%), while the median OS was not reached. In patients who achieved CR/PR per RECIST v1.1, the median duration of response was 10.3 months (95% CI, 8.2–13.9 months). The safety profiles are summarized in Supplementary Table 2. The most common adverse events were hand-foot skin reaction (n = 14, 33.3%), alopecia (n = 11, 26.2%), and skin rash (n = 9, 21.4%). Grade 3 adverse events were noted in 10 patients (23.8%), with the most frequent grade 3 adverse event being increased aspartate aminotransferase (n = 2, 4.8%). No grade 4 adverse event or treatment-related death occurred.

ctDNA analysis and survival outcomes

We performed ctDNA analysis in all 42 patients, revealing that the most frequently mutated gene was TP53 (69%), followed by CTNNB1 (encoding β-catenin) (26%) (Extended Data Fig. 2a). Alterations of genes involved in the Wnt/β-Catenin pathway were not associated with poor survival outcomes (Fig. 1d). Patients with alterations of genes involved in the PI3K/mTOR pathway showed a trend towards a favorable OS compared to those without such alterations (Extended Data Fig. 2b).

Enhanced expression of immune-related genes and increased frequency of classical monocytes upon regorafenib-nivolumab treatment

Next, we analyzed the early immune kinetics²² by performing scRNA-seq of PBMC samples collected at baseline (C1D1) and on-treatment (C1D15) (Fig. 2a). We focused on two clinical subgroups of patients: those who had disease progression at the first evaluation or who showed a progressive increase in the tumor burden (i.e., SD with increase in tumor size at the first evaluation and PD at the second evaluations, respectively) (early progressors, n = 14, marked as red box) vs. patients who showed a continuous decrease in tumor burden (i.e., PR or SD with decrease in tumor size), which lasted at least 10 months (long-term responders, n = 15, marked as blue box) (Fig. 1b). We first analyzed differentially expressed genes (DEGs) of whole PBMCs over regorafenib-nivolumab treatment. DEGs upregulated on C1D15 were associated with overall robust immune activation signals, including innate immune response and robust type I and II interferon (IFN) response (Fig. 2b,c). Importantly, this was more prominent among long-term responders.

Unsupervised clustering identified 13 distinct immune subsets (Fig. 2d and Extended Data Fig. 3a). The proportions of classical monocytes and proliferating lymphocytes significantly increased over regorafenib-nivolumab treatment in both clinical subgroups, while the proportion of intermediate monocytes increased only among early progressors (Fig. 2e, Extended Data Fig. 3b,c). Pearson correlation coefficient (PCC) analysis revealed that classical monocytes had the greatest difference in gene expression profiles between C1D1 and C1D15 (Fig. 2f and Extended Data Fig. 3d), indicating that classical monocytes underwent prominent transcriptomic reprogramming over the treatment. Importantly, this qualitative change was more prominent among long-term responders than in early progressors (Fig. 2f and Extended Data Fig. 3d). These results suggest that quantitative and qualitative changes of classical monocyte populations may be associated with the efficacy of regorafenib-nivolumab treatment.

Diversification of TCR repertoire and enrichment of genes representing ICI-responsiveness of CD8 ⁺ T cells in long-term responders

On-treatment CD8⁺ T cells exhibited upregulation of gene expression related to type I and II IFN response, which was more prominent among long-term responders, as well as downregulation of gene expressions related to pathways implicated in apoptosis (Extended Data Fig. 4a,b). Only among the long-term responders, treatment also significantly increased the diversity of TCR clones of CD8⁺ T cells as represented by increased inversed Simpson index (Fig. 3a), which is reportedly associated with favorable ICI response^23–25.

We performed unsupervised clustering of NK/T cells (Extended Data Fig. 4c,d) and analysis of proliferating lymphocytes, which predominantly comprised proliferating CD8⁺ T cells (Extended Data Fig. 4e,f). This yielded the identification of five distinct CD8⁺ T-cell subclusters: naïve CD8⁺ T cells; effector CD8⁺ T cells overexpressing genes encoding cytotoxic molecules (e.g., GZMH, GZMB, and PRF1); PDCD1⁺ CD8⁺ T cells overexpressing PDCD1 (encoding PD-1), GZMK and EOMES; MKI67⁺ proliferating CD8⁺ T cells; and innate-like effector CD8⁺ T cells (Fig. 3b). Among these CD8⁺ T cell subsets, the effector CD8⁺ and MKI67 proliferating CD8⁺ T cells exhibited significantly increased frequencies in the overall population throughout treatment, which were more prominent among long-term responders than early progressors (Fig. 3c and Extended Data Fig. 4g).

When receiving ICI-based treatment, a dynamic increase in the proportion of proliferating CD8⁺ T cells is reportedly associated with favorable clinical outcomes^22,26. We analyzed expression levels of the inflammatory signature associated with favorable clinical outcomes with nivolumab monotherapy in HCC¹⁹, and found that it was prominently enriched in MKI67⁺ proliferating CD8⁺ T cells among long-term responders but not early progressors (Fig. 3d, upper panel). We also measured the expression levels of a signature representing transcriptomic changes following nivolumab monotherapy among responders²⁴, and found an increasing trend in MKI67⁺ proliferating CD8⁺ T cells which was more prominent among long-term responders than in early progressors (Fig. 3d, lower panel). Also, MKI67⁺ proliferating CD8⁺ T cells were enriched for genes relevant to T-cell responses (Extended Data Fig. 4h). In particular, on C1D15, cytotoxicity-related genes were upregulated on MKI67⁺ proliferating CD8⁺ T cells, only among long-term responders (Fig. 3e). These findings were validated by multicolor flow-cytometry, which revealed significantly increased proportions of KI-67⁺ proliferating cells (Fig. 3f,g) and cells harboring cytotoxic granules (granzyme B⁺/perforin⁺) (Fig. 3h,i) among CD8⁺ T cells and PD-1⁺ CD8⁺ T cells, respectively, only in long-term responders but not in early progressors.

Taken together, our findings demonstrated that long-term responders showed diversification of the TCR repertoire, and a prominent increase in the relative abundance of proliferating CD8⁺ T cells enriched with signals related to ICI-responsiveness and cytotoxicity. These results suggest that efficacy outcomes may be related to qualitatively and qualitatively distinct reprogramming of CD8⁺ T cells following regorafenib-nivolumab treatment.

Preferential M1-directed polarization following regorafenib-nivolumab treatment in long-term responders

We then investigated monocyte subpopulations. Within these cell populations, on C1D15, we observed upregulation of genes involving IFN responses and antigen presentation, which was more prominent among long-term responders (Extended Data Fig. 5a,b). We identified five distinct monocyte clusters: S100⁺ early-activated classical monocytes; activated classical monocyte_1, overexpressing IL1B, FOSB, JUN, and IER2; activated classical monocyte_2, overexpressing LGASLS2 and IFI6; antigen-presenting indeterminate monocytes; and M2-skewed non-classical monocytes, overexpressing genes involved in immunosuppression, including CSF1R, SIGLEC10, and VSIR (encoding VISTA) (Fig. 4a). Within monocyte subpopulations, the relative abundances of these monocyte populations did not significantly change with treatment (Fig. 4b and Extended Data Fig. 5c).

Next, we performed a differentiation pseudotime ordering analysis, which revealed two distinct diverging trajectories—with the classical monocyte_2 subgroup being the bifurcation point, and one trajectory towards activated classical monocyte_1 (immunostimulatory M1-like branch), and the other towards M2 skewed non-classical monocytes (immunosuppressive M2-like branch) (Fig. 4c). Upon regorafenib-nivolumab treatment, we observed an increased proportion of M1-branch cells (towards activated classical monocyte_1) among total monocytes in long-term responders, but not in early progressors. On the other hand, the proportion of M2-branch cells (towards non-classical monocytes) did not significantly change with treatment in either clinical subgroup (Fig. 4d). At the bifurcation point, the expression levels of the M1-like gene signature²⁷ significantly increased, more prominently in long-term responders than in early progressors (Fig. 4e). These results indicate that favorable outcomes following regorafenib-nivolumab treatment were associated with a higher probability of monocytic cells polarizing towards M1-like features. On the other hand, the expression levels of tumor-associated macrophage (TAM) signatures²⁸ were significantly increased in M2-skewed non-classical monocytes from early progressors, but not in those from long-term responders (Fig. 4f). Multicolor flow-cytometry analysis revealed a significant increase in the relative abundance of CD14⁺ classical monocytes to CD14-low CD16-high non-classical monocytes, only among long-term responders (Fig. 4g and Extended Data Fig. 5d,e).

To further examine whether the transcriptomic changes of monophagocytic subsets were partly attributable to regorafenib, we examined regorafenib-responsive gene sets derived from various in vitro contexts in an HCC model¹³. Throughout regorafenib-nivolumab treatment, the expression levels of these regorafenib-induced signatures significantly increased in the classical monocyte subsets among long-term responders, while their expression levels only mildly changed, or even showed a decreasing trend, among early progressors (Fig. 4h). This was further corroborated by our finding that a gene signature representing the knock-out of CSF1R (targeted by regorafenib) was enriched in C1D15 classical monocyte subsets from only long-term responders, and not those from early progressors (Fig. 4i). These results suggest that regorafenib plays a role in driving the differential transcriptomic reprogramming of classical monocytes, possibly through blockade of the CSF1R pathway, in patients with distinct clinical outcomes.

Interaction between proliferating CD8 ⁺ T cells and classical monocytes through IFN-γ pathways in long-term responders

We then performed an interactome analysis to investigate the possibility of dynamic interactions between monocyte subpopulations and proliferating CD8⁺ T cells upon study treatment (Fig. 5a). We found that proliferating CD8⁺ T cells provided IFN-γ-related signals to the S100⁺ early-activated classical monocyte and activated classical monocyte_1 subsets among long-term responders. In contrast, among early progressors, we found no meaningful interaction between CD8⁺ T-cell subsets and the S100⁺ early-activated classical monocyte and activated classical monocyte_1 subsets. Rather, in the early progressors, innate-like CD8⁺ T cells interacted with the classical monocyte_1, intermediate monocyte, and non-classical monocyte subsets (Fig. 5b).

Taken together, these results suggest divergently differential CD8⁺ T-cell and monocyte responses to regorafenib-nivolumab treatment, according to distinct clinical outcomes. Moreover, it appears that these responses were induced by immune-pharmacodynamic effects of regorafenib and/or nivolumab, as well as their dynamic interactions.

Upregulation of TMEM176A/B on classical monocytes in early progressors

The clinical relevance of dynamic changes in classical monocytes prompted us to examine the DEGs in classical monocytes between early progressors vs. long-term responders on C1D15. We found that TMEM176A and TMEM176B were the most significantly upregulated genes in early progressors (vs. long-term responders) (Fig. 6a and Extended Data Fig. 6a). TMEM176B was particularly overexpressed in antigen-presenting indeterminate monocytes and M2-skewed non-classical monocytes from early progressors (Fig. 6b).

TMEM176B reportedly inhibits the inflammasome response and adversely affects responsiveness to ICI²⁹. Therefore, we further compared the expression levels of genes involved in inflammasome response, and found that NLRP3 and IL18 were prominently upregulated throughout regorafenib-nivolumab in long-term responders, which was not evident in early progressors (Fig. 6c). Accordingly, the expression levels of a gene set representing the NLRP3 inflammasome pathway were significantly increased among long-term responders throughout regorafenib-nivolumab treatment, but not in early progressors (Fig. 6d and Extended Data Fig. 6b). In addition, plasma IL1RA (reflecting inflammasome response³⁰) levels tended to increase after treatment (C2D1) only among long-term responders (Fig. 6e). These findings suggest that early disease progression may be associated with upregulation of TMEM176A/B, which may prevent an effective inflammasome response in this therapeutic context.

In this phase 2 study, we investigated the clinical outcomes and dynamic immune landscapes associated with regorafenib-nivolumab as first-line therapy in patients with unresectable HCC. The primary endpoint was met: the ORR was 31.0% with the median PFS of 7.38 months. These efficacy outcomes were comparable with the previously reported efficacy outcomes with other first-line anti-PD-1/PDL1-MKI combinations^5–7 or atezolizumab-bevacizumab³. Overall, regorafenib-nivolumab was well-tolerated with previously known toxicity profiles, and no new safety signal with the combination. These findings were consistent with a previous phase 1 study of regorafenib-nivolumab in other gastrointestinal cancers¹⁸. The present results suggest that regorafenib-nivolumab may be a clinically feasible first-line treatment option in patients with unresectable HCC, warranting further investigation.

In the current study, the systematic collection of samples in a clinical trial setting provided a unique opportunity to explore the immune landscapes reprogrammed by regorafenib-nivolumab (summarized in Supplementary Table 3). To our knowledge, this is the first study to explore, at a single-cell level, the dynamic immune landscapes of HCC patients treated with first-line ICI-based combination therapy. We revealed the clinical relevance of diverging CD8⁺ T-cell and classical monocyte responses, which could provide novel specific evidence for future development of biomarkers and novel immunotherapeutic strategies in HCC patients.

Our finding that the characteristics of CD8⁺ T cells upon regorafenib-nivolumab treatment were associated with efficacy outcomes solidifies the importance of T-cell activation in the context of ICI-based treatment. The TCR repertoire diversification in long-term responders following regorafenib-nivolumab treatment suggests that favorable outcomes were associated with the clonal expansion of CD8⁺ T cells recognizing a variety of neoantigens to a level that was detectable in peripheral blood^23–25. To our knowledge, this is the first such demonstration in HCC patients receiving ICI-based treatments. A proliferative burst of CD8⁺ T cells prominently seen among long-term responders accords well with previous studies dealing with its clinical value upon ICI-based treatment^22,26. We also found that only long-term responders exhibited enrichment of genes representing ICI responsiveness and cytotoxicity in proliferating MKI67⁺ CD8⁺ T cells, which further directed classical monocytes through IFN-γ-related pathways. This previously under-recognized clinically relevant interaction observed in long-term responders suggests that these interactions could potentially be harnessed to further enhance anti-tumor response, and suggests that early progressors may have the presence of factors precluding effective T-cell responses, and subsequent monocyte activation, despite regorafenib-nivolumab treatment.

Diverging monocyte response according to clinical outcomes is the one of the unique aspects of the present study. It is increasingly recognized that myeloid lineage cells play a critical role in meticulously balancing anti-tumor responses and immuno-suppression in the context of ICI-based treatments. In HCC patients, we previously found that the increased frequency of classical monocytes upon nivolumab monotherapy was more prominent in patients with durable clinical benefit than in those without³¹. CD38⁺ macrophages have also been associated with favorable survival outcomes in HCC patients treated with anti-PD-1/PD-L1³². Recent single-cell analysis demonstrated that the abundance of CD11c⁺ antigen-presenting cells was associated with the response to anti-PD-1 therapy in HCC patients³³. In the present study, we showed interactions between proliferating MKI67⁺ CD8⁺ T cells and classical monocyte subsets, as well as a higher probability of M1-directed polarization in long-term responders. These results point to the clinical relevance of positive associations between IFN-γ response and myeloid populations. Of note, our results validate the clinical relevance of the IFN-γ-directed myeloid response in the context of anti-PD-1/PD-L1-MKI combinations, which warrants the development of relevant myeloid-related biomarkers in this setting.

Conversely, in early progressors, the signature of immunosuppressive tumor-associated macrophages was prominently enriched on M2-skewed non-classical monocytes overexpressing immune suppressive markers, such as CSF1R, SIGLEC10, and VSIR (encoding VISTA), despite regorafenib-nivolumab treatment. This accords well with a previous report that immunosuppressive Arg-1-expressing CD163⁺ macrophages were enriched in HCC patients not responding to cabozantinib-nivolumab³⁴. To further examine these diverging monocyte responses in different clinical subgroups, we compared the gene expression profiles of monocyte subsets, which revealed that TMEM176A/B was associated with early disease progression. TMEM176A/B is a surface protein highly expressed on myeloid immune cells, which was recently shown to prevent inflammasome responses, thereby preventing effective ICI-mediated immune responses²⁹. Accordingly, regorafenib-nivolumab did not sufficiently increase the expression levels of NLRP3-related genes on monocytic populations from early progressors, possibly because of upregulation of TMEM176A/B. This novel association between early disease progression and up-regulation of TMEM176A/B coupled with an ineffective inflammasome response suggests a possibility to develop novel immunotherapeutic strategies to augment the inflammasome-related response, to overcome primary resistance to ICI-based treatments in HCC patients.

The clinical benefits of ICI-based combinations suggest different additive and/or synergistic effects of the combination partners. In a study of the atezolizumab-bevacizumab combination, the addition of bevacizumab to atezolizumab led to decreases in VEGFR2 expression levels and increased regulatory T cells, leading to improved survival outcomes particularly in HCC patients with high expression of VEGFR2, increased frequency of regulatory T cells, and enrichement of myeloid inflammation signatures²¹. On the other hand, in a study of neoadjuvant cabozantinib-nivolumab in HCC patients, the addition of cabozantinib decreased the level of CXCL1, a chemokine ligand involved in immune-resistance, and subsequently promoted T-cell activation³⁴. Cabozantinib also reportedly induces intratumoral neutrophil infiltration, which further enhances the inflammatory phenotype, when used in combination with nivolumab in preclinical models of HCC³⁵. In addition to its anti-angiogenic activity, regorafenib can potentially modulate myeloid cell populations by inhibiting the CSF-1/CSF-1R pathway^8,13, which drives the immunosuppressive gradient of myeloid cells. In our current study, we demonstrated regorafenib-specific effects on classical monocytes, specifically in long-term responders—which was characterized by enrichment of a regorafenib-induced gene signature and a gene signature representing CSF-1 deficient status, highlighting the clinical relevance of the immunomodulatory effect of regorafenib. The regorafenib-nivolumab combination is actively being explored in other cancer types (NCT04879368), and regorafenib plus anti-PD-1/L1 combinations are being investigated in HCC patients (NCT04718909, NCT04183088), supporting the clinical relevance and generalizability our present findings.

Alterations in the WNT/β-catenin pathway have been previously demonstrated to be associated with excluding anti-tumor immune responses, thereby conferring resistance to ICIs^36,37. However, Our ctDNA analysis revealed that genetic alterations in the Wnt/β-catenin pathway was not associated with poor survival outcomes. Similarly, the survival outcomes were comparable between those with versus without a CTNNB1 (encoding β-catenin) mutation, among HCC patients treated with atezolizumab plus bevacizumab²¹. The available data indicate that there is no need to preclude patients with WNT/β-catenin pathway mutations from receiving ICI-based combinations involving anti-angiogenic agents. Mutations in the PI3K/mTOR pathway detected by ctDNA are reportedly associated with poor survival outcomes in HCC patients treated with TKIs, but with a trend towards favorable survival outcomes in patients treated with ICI³⁸. In our analysis, patients with alterations in the PI3K/mTOR pathway showed a trend towards favorable OS compared to those without³⁸. Given the small numbers of patients involved in these results, further studies are warranted to delineate the clinical implications of mutations in the context of ICI-based combinations.

The absence of a control group (i.e., nivolumab monotherapy) may be one of the limitations of the current study, precluding the accurate interpretation on the effects of adding regorafenib to nivolumab. Nevertheless, given that anti-PD-1/PD-L1 monotherapy was not a standard 1st line option at the time of study design, we did not include a control group involving nivolumab monotherapy from the study design. Moreover, the subsequently reported results of the CheckMate459 which failed to demonstrate the superiority of nivolumab over sorafenib ¹, which also does not support the use of the nivolumab monotherapy control arm.

In conclusion, our present results demonstrate that regorafenib-nivolumab therapy has clinical activity and is well tolerated as a first-line treatment for patients with unresectable HCC. The differential quantitative and qualitative responses of CD8⁺ T cells and classical monocytes, and their interconnected relationships, upon regorafenib-nivolumab treatment highlight these cellular subsets as critical immune regulators in this therapeutic context. Modulation of the inflammasome response to overcome resistance to ICI-based treatments may be a potential novel immunotherapeutic approach that warrants further investigations in HCC patients.

HCC, hepatocellular carcinoma; ORR, objective response rate; OS, overall survival; ICI, immune checkpoint inhibitor; MKI, multi-kinase inhibitor; ctDNA, circulating tumor DNA; sc-RNA seq, single-cell RNA sequencing; TCR, T-cell receptor; ECOG, Eastern Cooperative Oncology Group; RECIST, Response Evaluation Criteria in Solid Tumors; NCI-CTCAE, National Cancer Institute Common Terminology Criteria for Adverse Events; PFS, progression-free survival; PBMC, peripheral blood mononuclear cell; PCA, principal component analysis; SNN, shared nearest neighbor; UMAP, uniform manifold approximation and projection; DEG, differentially expressed gene; GO.BP, Gene Ontology Biological process databases; PCC, Pearson correlation coefficient; CR, complete response; PR, partial response; CI, confidence interval; EP, early progressors; LR, long-term responders; IFN, interferon; TAM, tumor-associated macrophages.

Disclosures:

CY received honoraria from Servier, Bayer, AstraZeneca, Merck Sharp & Dohme, Eisai, Celgene, Bristol Myers Squibb, Ipsen, Novartis, Boryung Pharmaceuticals, Mundipharma, and Roche; and received research grants from Servier, Bayer, AstraZeneca, Ono Pharmaceuticals, Ipsen, and Boryung Pharmaceuticals. HDK received honoraria from AstraZeneca, Bristol Myers Squibb, Ono Pharmaceuticals, Boryung Pharmaceuticals, and Boostimmune. RSF reports grants or contracts to their institution from Adaptimmune, Bayer, Bristol Myers Squibb, Eisai, Eli Lilly, Pfizer, Roche, and Genentech; consulting fees to themself from Merck, AstraZeneca, Bayer, Bristol Myers Squibb, Exelixis, Cstone, Hengrui, Eisai, Eli Lilly, MSD, Pfizer, Roche, and Genentech; payment or honoraria to themself from Genentech; and participation on a data safety monitoring or advisory board from AstraZeneca, and Hengrui.

Funding

This study was funded in part by Bayer, ONO Pharmaceuticals, Genome Insight Inc., and Korean Cancer Study Group (grant number: KCSG HB20-08).

Yau, T. et al. Nivolumab versus sorafenib in advanced hepatocellular carcinoma (CheckMate 459): a randomised, multicentre, open-label, phase 3 trial. The Lancet. Oncology 23, 77-90, doi:10.1016/s1470-2045(21)00604-5 (2022).
Finn, R. S. et al. Pembrolizumab As Second-Line Therapy in Patients With Advanced Hepatocellular Carcinoma in KEYNOTE-240: A Randomized, Double-Blind, Phase III Trial. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 38, 193-202, doi:10.1200/jco.19.01307 (2020).
Finn, R. S. et al. Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. The New England journal of medicine 382, 1894-1905, doi:10.1056/NEJMoa1915745 (2020).
Abou-Alfa, G. K. et al. Tremelimumab plus Durvalumab in Unresectable Hepatocellular Carcinoma. NEJM Evidence 1, EVIDoa2100070, doi:doi:10.1056/EVIDoa2100070 (2022).
Qin, S. et al. LBA35 camrelizumab (C) plus rivoceranib (R) vs. sorafenib (S) as first-line therapy for unresectable hepatocellular carcinoma (uHCC): a randomized, phase III trial. Annals of Oncology 33, S1401-S1402 (2022).
Kelley, R. K. et al. Cabozantinib plus atezolizumab versus sorafenib for advanced hepatocellular carcinoma (COSMIC-312): a multicentre, open-label, randomised, phase 3 trial. The Lancet. Oncology 23, 995-1008, doi:10.1016/s1470-2045(22)00326-6 (2022).
Finn, R. et al. LBA34 primary results from the phase III LEAP-002 study: lenvatinib plus pembrolizumab versus lenvatinib as first-line (1L) therapy for advanced hepatocellular carcinoma (aHCC). Annals of Oncology 33, S1401 (2022).
Wilhelm, S. M. et al. Regorafenib (BAY 73-4506): a new oral multikinase inhibitor of angiogenic, stromal and oncogenic receptor tyrosine kinases with potent preclinical antitumor activity. International journal of cancer 129, 245-255, doi:10.1002/ijc.25864 (2011).
Bruix, J. et al. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet (London, England) 389, 56-66, doi:10.1016/s0140-6736(16)32453-9 (2017).
Yoo, C. et al. Regorafenib in previously treated advanced hepatocellular carcinoma: Impact of prior immunotherapy and adverse events. Liver international : official journal of the International Association for the Study of the Liver 40, 2263-2271, doi:10.1111/liv.14496 (2020).
Bronte, V. & Murray, P. J. J. N. m. Understanding local macrophage phenotypes in disease: modulating macrophage function to treat cancer. Nature medicine 21, 117-119 (2015).
Cannarile, M. A. et al. Colony-stimulating factor 1 receptor (CSF1R) inhibitors in cancer therapy. Journal for immunotherapy of cancer 5, 53, doi:10.1186/s40425-017-0257-y (2017).
Ou, D. L. et al. Regorafenib enhances antitumor immunity via inhibition of p38 kinase/Creb1/Klf4 axis in tumor-associated macrophages. Journal for immunotherapy of cancer 9, doi:10.1136/jitc-2020-001657 (2021).
Sharma, A. et al. Onco-fetal Reprogramming of Endothelial Cells Drives Immunosuppressive Macrophages in Hepatocellular Carcinoma. Cell 183, 377-394.e321, doi:10.1016/j.cell.2020.08.040 (2020).
Nguyen, P. H. D. et al. Trajectory of immune evasion and cancer progression in hepatocellular carcinoma. Nature communications 13, 1441, doi:10.1038/s41467-022-29122-w (2022).
Wu, C. et al. Myeloid signature reveals immune contexture and predicts the prognosis of hepatocellular carcinoma. The Journal of clinical investigation 130, 4679-4693, doi:10.1172/jci135048 (2020).
Peranzoni, E., Ingangi, V., Masetto, E., Pinton, L. & Marigo, I. Myeloid Cells as Clinical Biomarkers for Immune Checkpoint Blockade. Frontiers in immunology 11, 1590, doi:10.3389/fimmu.2020.01590 (2020).
Fukuoka, S. et al. Regorafenib Plus Nivolumab in Patients With Advanced Gastric or Colorectal Cancer: An Open-Label, Dose-Escalation, and Dose-Expansion Phase Ib Trial (REGONIVO, EPOC1603). Journal of clinical oncology : official journal of the American Society of Clinical Oncology 38, 2053-2061, doi:10.1200/jco.19.03296 (2020).
Sangro, B. et al. Association of inflammatory biomarkers with clinical outcomes in nivolumab-treated patients with advanced hepatocellular carcinoma. Journal of hepatology 73, 1460-1469, doi:10.1016/j.jhep.2020.07.026 (2020).
Neely, J. et al. Genomic and transcriptomic analyses related to the clinical efficacy of first-line nivolumab in advanced hepatocellular carcinoma from the phase 3 CheckMate 459 trial. Cancer Res 82, 2145-2145 (2022).
Zhu, A. X. et al. Molecular correlates of clinical response and resistance to atezolizumab in combination with bevacizumab in advanced hepatocellular carcinoma. Nature medicine 28, 1599-1611, doi:10.1038/s41591-022-01868-2 (2022).
Kim, K. H. et al. The First-week Proliferative Response of Peripheral Blood PD-1(+)CD8(+) T Cells Predicts the Response to Anti-PD-1 Therapy in Solid Tumors. Clinical cancer research : an official journal of the American Association for Cancer Research 25, 2144-2154, doi:10.1158/1078-0432.Ccr-18-1449 (2019).
Yost, K. E. et al. Clonal replacement of tumor-specific T cells following PD-1 blockade. Nature medicine 25, 1251-1259, doi:10.1038/s41591-019-0522-3 (2019).
Au, L. et al. Determinants of anti-PD-1 response and resistance in clear cell renal cell carcinoma. Cancer cell 39, 1497-1518.e1411, doi:10.1016/j.ccell.2021.10.001 (2021).
Han, J. et al. TCR Repertoire Diversity of Peripheral PD-1(+)CD8(+) T Cells Predicts Clinical Outcomes after Immunotherapy in Patients with Non-Small Cell Lung Cancer. Cancer immunology research 8, 146-154, doi:10.1158/2326-6066.Cir-19-0398 (2020).
Huang, A. C. et al. T-cell invigoration to tumour burden ratio associated with anti-PD-1 response. Nature 545, 60-65, doi:10.1038/nature22079 (2017).
Cassetta, L. et al. Human Tumor-Associated Macrophage and Monocyte Transcriptional Landscapes Reveal Cancer-Specific Reprogramming, Biomarkers, and Therapeutic Targets. Cancer cell 35, 588-602.e510, doi:10.1016/j.ccell.2019.02.009 (2019).
Azizi, E. et al. Single-Cell Map of Diverse Immune Phenotypes in the Breast Tumor Microenvironment. Cell 174, 1293-1308.e1236, doi:10.1016/j.cell.2018.05.060 (2018).
Segovia, M. et al. Targeting TMEM176B Enhances Antitumor Immunity and Augments the Efficacy of Immune Checkpoint Blockers by Unleashing Inflammasome Activation. Cancer cell 35, 767-781.e766, doi:10.1016/j.ccell.2019.04.003 (2019).
Chan, A. H. & Schroder, K. Inflammasome signaling and regulation of interleukin-1 family cytokines. The Journal of experimental medicine 217, doi:10.1084/jem.20190314 (2020).
Jeon, S. H. et al. Dynamic changes in peripheral blood monocytes early after anti-PD-1 therapy predict clinical outcomes in hepatocellular carcinoma. Cancer immunology, immunotherapy : CII, 1-14, doi:10.1007/s00262-022-03258-6 (2022).
Ng, H. H. M. et al. Immunohistochemical scoring of CD38 in the tumor microenvironment predicts responsiveness to anti-PD-1/PD-L1 immunotherapy in hepatocellular carcinoma. Journal for immunotherapy of cancer 8, doi:10.1136/jitc-2020-000987 (2020).
Chuah, S. et al. Uncoupling immune trajectories of response and adverse events from anti-PD-1 immunotherapy in hepatocellular carcinoma. Journal of hepatology 77, 683-694, doi:10.1016/j.jhep.2022.03.039 (2022).
Ho, W. J. et al. Neoadjuvant Cabozantinib and Nivolumab Converts Locally Advanced HCC into Resectable Disease with Enhanced Antitumor Immunity. Nature cancer 2, 891-903, doi:10.1038/s43018-021-00234-4 (2021).
Esteban-Fabró, R. et al. Cabozantinib Enhances Anti-PD1 Activity and Elicits a Neutrophil-Based Immune Response in Hepatocellular Carcinoma. Clinical cancer research : an official journal of the American Association for Cancer Research 28, 2449-2460, doi:10.1158/1078-0432.Ccr-21-2517 (2022).
Pinyol, R., Sia, D. & Llovet, J. M. Immune Exclusion-Wnt/CTNNB1 Class Predicts Resistance to Immunotherapies in HCC. Clinical cancer research : an official journal of the American Association for Cancer Research 25, 2021-2023, doi:10.1158/1078-0432.Ccr-18-3778 (2019).
Spranger, S., Bao, R. & Gajewski, T. F. Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity. Nature 523, 231-235, doi:10.1038/nature14404 (2015).
von Felden, J. et al. Mutations in circulating tumor DNA predict primary resistance to systemic therapies in advanced hepatocellular carcinoma. Oncogene 40, 140-151, doi:10.1038/s41388-020-01519-1 (2021).

Yes there is potential Competing Interest. Changhoon Yoo received honoraria from Servier, Bayer, AstraZeneca, Merck Sharp & Dohme, Eisai, Celgene, Bristol Myers Squibb, Ipsen, Novartis, Boryung Pharmaceuticals, Mundipharma, and Roche; and received research grants from Servier, Bayer, AstraZeneca, Ono Pharmaceuticals, Ipsen, and Boryung Pharmaceuticals. Hyung-Don Kim received honoraria from AstraZeneca, Bristol Myers Squibb, Ono Pharmaceuticals, Boryung Pharmaceuticals, and Boostimmune. Richard S. Finn reports grants or contracts to their institution from Adaptimmune, Bayer, Bristol Myers Squibb, Eisai, Eli Lilly, Pfizer, Roche, and Genentech; consulting fees to themself from Merck, AstraZeneca, Bayer, Bristol Myers Squibb, Exelixis, Cstone, Hengrui, Eisai, Eli Lilly, MSD, Pfizer, Roche, and Genentech; payment or honoraria to themself from Genentech; and participation on a data safety monitoring or advisory board from AstraZeneca, and Hengrui.

Download PDF

Journal Publication

published 19 Feb, 2024

Read the published version in Nature Medicine →

Version 1

posted

You are reading this latest preprint version

Differential T-cell and monocyte responses in hepatocellular carcinoma treated with regorafenib plus nivolumab: an integrated clinical and biomarker analysis of the phase 2 RENOBATE trial

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Clinical outcomes of regorafenib-nivolumab

ctDNA analysis and survival outcomes

Enhanced expression of immune-related genes and increased frequency of classical monocytes upon regorafenib-nivolumab treatment

Preferential M1-directed polarization following regorafenib-nivolumab treatment in long-term responders

Upregulation of TMEM176A/B on classical monocytes in early progressors

Discussion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1