Highlights in Cancer Research: September 2024

Late relapse remains a major clinical challenge in estrogen receptor–positive breast cancer (ER+ BC) patients treated with adjuvant endocrine therapies (ETs). Despite the effectiveness of ETs in delaying relapse by targeting subclinical micrometastases, up to 50% of patients experience relapse through unknown mechanisms likely involving dormancy.

.
This study investigated the genetic and transcriptional changes behind tumour awakening by analysing the genomic landscape of late relapse and longitudinally profiling a rare cohort treated with long-term neoadjuvant ET. Complementing these clinical observations, the authors conducted an in vitro evolutionary study in unperturbed settings to record the adaptive strategies of individual cancer lineages along dormancy and awakening.

The data revealed that ER+ BC cells exposed to ETs enter dormancy in a stochastic pattern through epigenetic reprogramming. Notably, dormant cells can awaken unpredictably, exhibiting divergent phenotypes, and this reactivation doesn’t involve recurrent genetic alterations. Importantly, targeting the epigenome of dormant cells shows promise in combating adaptive resistance to ETs.

.
This study highlights the critical role of epigenetic adaptation in the evolution of resistance to ETs and emphasises the potential for developing therapeutic strategies aimed at targeting dormant cancer cells.

The tumor suppressor p53 is the most studied protein (gene) in biology. It is the most frequently mutated protein in cancer, producing thousands of diverse and inactivated mutants. Pharmacological targeting mutant p53 requires a strategy that challengingly restores rather than traditionally inhibits protein function. Over the past decades, approximately 70-100 research groups have reported the identification of small-molecule compounds capable of restoring the tumor-suppressive function to p53 mutants, and 7 of these compounds have entered 23 phase I-III clinical trials covering over 1000 cancer patients.

.
In this article, the authors employed 10 widely used p53 assays and evaluated 14 representative generic mutant p53 rescue compounds side-by-side (including all of the 6 commercially available compounds that have entered the clinical trials). Thousands of independent biological samples were generated in the evaluation. However, with the exception of ATO and its analogue PAT, they did not detect any reliable rescue in any of the 10 assays for the evaluated compounds. This result recapitulates the findings in the Cancer Cell paper that first reported rescue of mutant p53 by ATO in 2021.

Cancer-associated fibroblasts (CAFs) orchestrate an immunosuppressive tumour microenvironment (TME), significantly contributing to tumorigenesis and progression. In the article by Varveri et al, the authors identified α-SMA+ CAFs to be capable of processing and presenting tumour-associated antigens on T regulatory cells, through the formation of immunological synapses. These led to Treg cell activation and enhanced Treg-mediated immune suppression within the TME. Importantly, CAFs exhibited heightened autophagic activity which not only sustains them but also facilitated antigen-specific Treg activation. This process resulted in an amplified immunosuppressive milieu, promoting tumor immune evasion.

Cancer cells have the ability to hijack various cellular processes and hence despite presenting similar clinical characteristics tumours may present molecular differences with serious implications for disease prognostic and treatment response. The germline genome is highly variable, with differences across individuals occurring at millions of polymorphic sites. This variability may play a role in dictating the somatic evolution of the tumour. Mounting evidence suggests that a subset of T cells may respond to germline-derived epitopes during tumorigenesis.

In this paper, Houlahan et al investigated the role that germline variability may play in the development of various types of breast cancer by mediating immunoediting. Starting from the hypothesis that during tumour progression the body would select against genomic amplifications of germline variants that produce MHC class I antigenic determinants, the authors show that germline-derived epitopes have the ability to influence the molecular subtype that breast tumours commit to. The authors of this paper suggest that this germline-derived epitope burden negative selection is likely to occur beginning from precancerous lesions and persists throughout tumour progression influencing the metastatic predisposition. When tumours develop despite a high germline-derived epitope burden, they develop immune evasion mechanisms and hence tend to be more aggressive.

The transition from healthy to cancerous tissue is a complex and still poorly understood
process. This limited understanding of cancer initiation is largely due to the difficulty of modeling tumorigenesis in the laboratory. Although conventional cancer models are useful to study simple cancer cell behaviors, they lack the cellular diversity, tissue organization, longevity, and experimental versatility that are needed to capture the intricacies of tumorigenesis.

.
To solve this problem, we have developed a cancer model that can capture tumorigenesis ex vivo. We have used advanced microfabrication, tissue engineering, and optogenetic techniques to achieve this. Specifically, we have genetically engineered healthy colon epithelial cells to acquire cancerous mutations when exposed to blue light. These inducible cells have then been cultured in a microfluidic device designed to support the formation of long-lived miniature colon tissues (“mini-colons”) with in vivo-like structure.

.
This combination of techniques has enabled us to trigger oncogenic mutations in specific regions of “lab-made” colon epithelia, which we could then track in real-time and at single-cell resolution. This has enabled us to follow the progression of mutated cells from early-stage hyperplasias to full-blown tumors, capturing the tumorigenic process in
unprecedented detail.
.

The presence of metastasis to distal tissues is a main indicator of prognosis for cancer patients. Metastatic subclones possess certain characteristics such as increased mobility and the ability to stay dormant over long period of times, which underly their ability to seed distal organs. However metastatic subclones often are not particularly important in the growth of the primary tumour, making it difficult to identify them. Moreover, several reports highlighted the ability of cancer cells to further disseminate from metastatic sites to other organs and back to the primary tumours in a metastatic cascade. Understanding the kinetics of cancer cell dissemination could help develop novel therapeutic strategies targeting mechanisms of metastatic spread.

In this paper, Serio et al. are looking at the patterns of metastatic spread using a mouse model for an aggressive metastatic form of prostate cancer genetically induced through PTEN/TP53 deletion. Using CRISPR/Cas9-based barcoding technology the authors determined the migration histories of cancer cells from several metastatic sites, corresponding to human prostate cancer topologies. The authors observed a high degree of clonal heterogeneity from the primary tumour in the metastatic seeding, however, only a few clones showed the ability to invade. Most metastasis observed originated from the primary tumour, with secondary seeding being an even more infrequent event, and re-seeding of the primary tumour being an exceptional occurrence. The authors observed a widespread distribution of the initiator clones between metastatic sites suggesting polyclonal metastatic seeding might occur early in tumorigenesis. The mutations driving tumour formation are likely to impact the adaptations that cells undergo during the metastatic process hence altering the trajectories of metastatic spread. Therefore, future work should focus on investigating different starting gene combinations.

Tumor growth mirrors aspects of fetal development, particularly in the adaptability of cells and how tissues expand. In previous research, the authors identified similarities between the tumor microenvironment (TME) in hepatocellular carcinoma (HCC) and fetal liver development. They referred to the process where non-cancerous cells, such as FOLR2+ tumor-associated macrophages (TAM) and PLVAP+ endothelial cells (EC) in the TME, undergo reprogramming similar to fetal development as ‘oncofetal reprogramming’.

In this article, the authors identified five subpopulations of cancer-associated fibroblasts (CAF) in HCC, among which the POSTN+ extracellular matrix (EM) CAF shared significant transcriptomic similarities to fibroblasts in the human fetal liver. Interactome analysis and spatial transcriptomics data revealed the co-localization and close crosstalk among the three oncofetal cell types, POSTN+ CAF, PLVAP+ EC, and FOLR2+ TAM. Furthermore, patients with a higher enrichment of oncofetal components in pre-treatment HCC tumors were more likely to experience early relapse after surgical resection compared to those with a less pronounced oncofetal ecosystem. Notably, the oncofetal signatures differentiated tumor cells in HCC patients into two distinct relapse patterns: one dominated by EMT-like reprogramming and the other characterized by an upregulated lipid metabolism program. Lastly, this study established connections between an enriched oncofetal ecosystem, elevated atezolizumab (anti-PD-L1) + bevacizumab (anti-VEGF) response signature (ABRS), increased Treg signature, and improved progression-free survival.

Breast cancer cell (BCC) metastasis to the leptomeninges (LM), the cerebrospinal fluid (CSF)-containing membranes surrounding the brain and spinal cord, is a rapidly fatal disease complication. Despite recent advances in treating brain parenchymal metastasis, standard of care for LM disease (LMD) has remained essentially unchanged for decades. The paucity of targeted molecular therapies to treat LMD is attributed to our poor understanding of the molecular mechanisms governing LM metastasis and the narrow toxicity window of the nervous system.

In this article, the authors show that bone-metastatic BCCs can bypass the blood-brain and blood-CSF barriers and invade the LM along the abluminal surface of emissary vessels connecting the bone marrow to the central nervous system. This process, dependent on BCC α6 integrin-laminin interactions with the vascular basement membrane, is a novel and efficient route of solid tumor LM metastasis. Once in the LM, BCCs are exposed to a tissue niche that is at once a harsh environment for tumor growth due to relative nutrient deprivation and a “sanctuary site” of therapeutic resistance. Little is understood about the role the immune system plays in disease control or tumor promotion in the LM. The authors’ new data shed light on these interactions. Using intravital microscopy, they show that BCCs in the LM are encased by perivascular macrophages that secrete the protective neurotrophin, GDNF, activating NCAM-dependent cell survival pathways in BCCs. This process mimics GDNF/macrophage developmental and neuroprotective pathways and highlights the cunning mechanisms tumors cells can invoke to metastasize successfully.

Bone is the most frequent site of breast cancer metastasis leading to severe morbidities. Although immunotherapies have revolutionized the treatment of several cancer types, their efficacy in treating breast cancer metastasis is still limited, and bone metastasis remain incurable. This study profiled the changes in the immune microenvironment during bone metastatic progression and characterized the crosstalk between immunosuppressive granulocytes (myeloid-derived suppressive cells; MDSCs) and dysfunctional cytotoxic T lymphocytes (CTLs).

.
Using an immunocompetent mouse model of spontaneous breast cancer bone metastasis, the authors found that bone metastatic lesions are infiltrated by MDSCs and PD-1 and TIGIT-expressing dysfunctional CTLs. Temporal transcriptome analysis revealed extensive communication between MDSCs and T cells via their expression of immune checkpoint molecules. The authors further identified IL1b as a key driver of immune suppression in granulocytes: targeting IL1b in vivo by neutralizing antibodies, or by BM transplantation from IL1b KO mice reduced MDSC accumulation and inhibited their immunosuppressive phenotype, restoring T cell killing activity. Importantly, TIGIT was shown to be expressed in human bone metastasis from various cancer types.

.
Finally, co-targeting IL1b and TIGIT reactivated anti-tumor immunity, and enhanced survival in mice with bone metastasis, suggesting that combinatorial targeting of IL1b and TIGIT may be a novel approach to treat bone metastasis.

KRAS mutations occur in approximately 30% of all human cancers, and frequently in pancreas, colon and lung tumors. Small molecule inhibitors that specifically target the inactive (OFF) state of KRAS G12C mutant proteins have demonstrated clinical efficacy in NSCLC, yet patients eventually progress and treatment resistant tumors frequently harbor non-G12C RAS mutations or other RAS pathway mutations. Notably, there are no approved RAS-targeted therapy options for the majority of patients with non-G12C RAS mutant cancers, in particular patients with PDAC of whom over 90% have RAS mutant tumors. Therefore, there is a significant unmet medical need for a direct RAS inhibitor that targets multiple oncogenic RAS mutations as a potential therapeutic strategy for KRAS addicted cancers.

.
RMC-7977 is a potent RAS(ON) multi-selective, noncovalent tri-complex small molecule
inhibitor that is selective for the active, GTP-bound state of both mutant and wild-type variants of the canonical RAS isoforms (KRAS, NRAS and HRAS). RMC-7977 forms a tri-complex with RAS(ON) and cyclophilin A (CypA), disrupting RAS effector binding through steric occlusion. RMC-7977 binds to the switch I effector binding domain and occupies a binding pocket between RAS(ON) and CypA. This pocket leaves a groove containing common oncogenic mutational hotspots G12, G13 and Q61 unoccupied, providing a structural basis for tri-complex formation with, and inhibition of, multiple RAS-GTP variants by RMC-7977.

RMC-7977 potently inhibits proliferation in RAS-addicted cancer cells in vitro. Consistent with this mechanism of action, in a large panel of >800 cancer cell lines, KRAS G12X and NRAS Q61X mutant cancer cell lines were associated with RMC-7977 sensitivity while BRAF V600E mutant cells correlated with resistance. In vivo, oral administration of RMC-7977 potently inhibited RAS pathway activation in tumors and induced tumor regressions in multiple KRAS mutant human xenograft models in mice. Preclinical data also demonstrate that RMC-7977 has the potential to overcome many of the resistance mechanisms observed clinically for KRASG12C(OFF) inhibitors.