To Quiesce or Senesce, That Is the Question for Dedifferentiated Liposarcoma

In this issue of Clinical Cancer Research, Gleason and colleagues define new hallmarks of cellular senescence in dedifferentiated liposarcoma (DDLPS; ref. 1). As DDLPS is defined by an amplification on chromosome 12q of CDK4 and MDM2, there has been much work to use this as a therapeutic target for treatment using either CKD4 inhibitors or MDM2 inhibitors (2, 3). In the current work, the authors demonstrated that geroconversion, the transition of quiescent cells to senescent cells, is an important step in understanding the efficacy of CDK4 inhibition and it explains why some patients fail to benefit from them.

DDLPS is adipocytic sarcoma that commonly arise in the retroperitoneum or extremities (2). In the advanced setting, they are aggressive with relative resistance to chemotherapy. This has led to great interest in targeting both CDK4 and MDM2 in advanced DDLPS. There have been several trials evaluating CDK4 inhibitors in DDLPS (4, 5). Dickson and colleagues evaluated patients in two phase II trials using palbociclib, first at 200 mg daily for 14 days in a 21 day cycle, and later at 125 mg daily for 21 days in a 28 day cycle. Both trials noted a progression-free survival (PFS) of 18 weeks (4, 5). A phase II trial by Dickson and colleagues treated 29 patients with the CDK4 inhibitor abemaciclib 200 mg BID continuously. They observed a 12-week PFS of 76% and median PFS of 30.4 weeks (6). This led to the phase III SARC041 trial, a randomized double-blind study of abemaciclib versus placebo in patients with advanced DDLPS which is currently accruing (7). There has also been interest in targeting the amplification of MDM2 in DDLPS (8–11). MANTRA is a phase III trial comparing the MDM2 inhibitor milademetan versus Trabectedin in DDLPS. PFS for milademetan was 3.6 months compared with Trabectedin's 2.2 months; however, this trial did not meet its primary endpoint (9). A phase I trial of the MDM2 inhibitor Idasanutlin by Italiano and colleagues, noted stable disease in 31% (26/85) of patients, two with prolonged stability at > 600 days (10). The phase Ia/Ib dose escalation/expansion study of MDM2-p53 antagonist BI907828 demonstrated a manageable safety profile. In 42 patients with DDLPS there was a 19% (8/42) overall response rate and 88.1% disease control rate (11). BI907828 is also currently being evaluated versus doxorubicin as first line in patients with DDLPS (NCT05218499).

Senescence is a complex cellular event leading to stable cell-cycle exit in response to CDK4/6 inhibitors (CDKi) such as palbociclib and abemaciclib. Previous work by Koff and colleagues laid the foundation for this understanding, establishing the important difference between irreversible cell-cycle exit seen in responders to CDKis and reversible cell-cycle exit (quiescence) in nonresponders (12). The experiments in this work established some of the key molecular features that distinguish between these two cellular states. One of the important hypotheses born from this data include MDM2 regulation as a critical factor in inducing senescence machinery. Knockdown of MDM2 alone was sufficient to induce senescence even in nonresponder cell lines. The data additionally suggest that ATRX may be directly regulating MDM2 expression as its loss prevents MDM2 turnover and thus, senescence. The authors also importantly show the mechanism for geroconversion is p53 independent, explaining why there might be limited therapeutic efficacy in the combination of an MDM2 inhibitor (which liberates p53) and a CDKi (12). Further preclinical data by Sriraman and colleagues supports this, by showing that CDK4 inhibition attenuates the p53 response to MDM2-targeted drugs, resulting in decreased efficacy (13).

Gleason and colleagues (1) build on this information in several ways (Fig. 1). Firstly, they aim to better define the molecular markers of senescence by establishing that it is a time-mediated process. Whereas the work by Koff and colleagues measured levels of senescence-associated proteins such as beta-galactosidase (SA-β-gal) and HP1ϒ to define the senescent state, this work demonstrates that true senescence (the point at which cells could no longer divide when the CDKi was removed from cellular media) occurs long after SA-β-gal levels had peaked. Attempting to identify the genomic regulators of geroconversion and thus, a more specific marker of senescence led to the important discovery of ANGPTL4 as late phase gene product necessary for induction of the senescence-associated secretory phenotype. ANGPTL4 proved to be a better marker of senescence because it demonstrated low expression in untreated and quiescent cells and elevated expression in CDK4i treated cells that underwent senescence. This discovery helps further the understanding of the cellular events surrounding stable cell-cycle exit and what might differ between responders to CKI4s and nonresponders.

Furthermore, this work delves into the immunological changes in response to CDK4i treatment. Specifically, the effect of cellular immune cell infiltration on senescence and ultimately patient response. The authors show that senescence is correlated with an increase in TILs, specifically CD4⁺ cells. Examining paired biopsy samples from baseline and at 1 month on abemaciclib in participants of the ongoing phase III SARC41 clinical trial, the authors show no clear correlation between the presence or absence of immune inflammatory cells and patient outcome. Interestingly, they do note that patients with increased TILs and senescence markers had better initial outcomes at 1 month, however had a 60% greater risk of progression as compared with patients who showed neither of those responses at the first biopsy. The authors hypothesize that perhaps a chronic inflammatory state may create a permissive environment for resistance. It further suggests a strategy in which senolytics are combined with CDKis to improve long-term outcomes that will be studied in clinical trials.

The long-term work of the Koff laboratory and the clinical investigators has demonstrated that perseverance leads to success. The pursuit of CDK4 inhibition for DDLPS is now over 10 years old and in phase III clinical trials. It is through careful work combining clinical investigation with mechanistic discovery that the new understanding quiescence versus senescence was identified. This is a new hallmark of DDLPS and will help to define the next studies of CDK4i after the completion of SARC041. As such, to quiesce or senesce, that is the question for DDLPS.

B.A. Van Tine reports grants and other support from Polaris; personal fees from Bayer, Deciphera Pharmaceuticals, Daiichi Sankyo, EcoR1, Advenchen, Putnam, Salarius Pharmaceuticals, Boxer Capital, Acuta Capital Partners, Aadi, Race Oncology, Hinge Bio, Kronos Bio, Sonata Therapeutics, Iterion Therapeutics, Total Health Conference, Apexigen, PTC Therapeutics, Boehringer Ingelheim, Agenus, Regeneron Pharmaceuticals, and Curis; and other support from Adaptimmune outside the submitted work. No disclosures were reported by the other authors.