underlies some cases of CDK4/6 inhibitor resist-
ance (see below) ( 111 ), stimulation of lysosomo-
genesis by CDK4/6 inhibitors might limit
their clinical efficacy by inducing resistance.
Lastly, CDK4/6 inhibition impaired lysoso-
mal function and the autophagic flux in cancer
cells. It was argued that this lysosomal dys-
function was responsible for the senescent phe-
notype in CDK4/6 inhibitor–treated cells ( 110 ).
Because lysosomes are essential for autophagy,
the authors co-treated TNBC xenografts with
abemaciclib plus an AMPK activator, A769662
(which induces autophagy), and found that
this led to cancer cell death and subsequent
regression of tumors ( 110 ).
Cyclin D3–CDK6 phosphorylates and in-
hibits two rate-limiting glycolytic enzymes,
6-phosphofructokinase and pyruvate kinase
M2. This redirects glycolytic intermediates into
the pentose phosphate pathway (PPP) and
serine synthesis pathway. Through this mech-
anism, cyclin D3–CDK6 promotes the pro-
duction of reduced nicotinamide adenine
dinucleotide phosphate (NADPH) and reduced
glutathione (GSH) and helps to neutralize ROS
( 112 ). Treatment of tumors expressing high
levels of cyclin D3–CDK6 (such as leukemias)
with CDK4/6 inhibitors reduced the PPP- and
serine-synthesis pathway flow, thereby deplet-
ing the antioxidants NADPH and GSH. This
increased ROS levels and triggered tumor cell
apoptosis ( 112 ).
Another link between cyclin D–CDK4/6 in
metabolism and cancer was provided by the
observation that livers of obese/diabetic mice
up-regulate cyclin D1 expression ( 113 ). Treat-
ment of these mice with an antidiabetic com-
pound, metformin, reduced liver cyclin D1 levels
and largely protected mice against develop-
ment of hepatocellular carcinoma. Also, genetic
ablation of cyclin D1 protected obese/diabetic
mice from liver cancer, and administration of
palbociclib inhibited liver cancer progression.
These treatments had no effect on tumors in
nonobese animals ( 113 ). These observations
raise the possibility of using antidiabetic com-
pounds with CDK4/6 inhibitors for treatment
of liver cancers in obese patients.
CDK4/6 inhibitors and antitumor
immune responses
Several recent reports have started to unravel
how inhibition of CDK4/6 influences anti-
tumor immune responses, acting both on tu-
mor cells as well as on the tumor immune
environment (Fig. 2B). Treatment of breast
cancer–bearing mice or breast cancer cells
with abemaciclib activated expression of endo-
genous retroviral elements in tumor cells,
thereby increasing the levels of double-stranded
RNA. This, in turn, stimulated production of
type III interferons and increased presentation
of tumor antigens. Hence, CDK4/6 inhibitors,
by inducing viral gene expression, trigger anti-
viral immune responses that help to eliminate
the tumor ( 114 ).
Inhibition of CDK4/6 also affects the im-
mune system by impeding the proliferation of
CD4+FOXP3+regulatory T cells (Tregs), which
normally inhibit the antitumor response. Be-
cause cytotoxic CD8+T cells are less affected
by CDK4/6 inhibition, abemaciclib treatment
decreases the Treg/CD8+ratio of intratumoral
T cells and facilitates tumor cell killing by
cytotoxic CD8+T cells ( 114 ).
Inhibition of CDK4/6 also resulted in activ-
ationofTcellsthroughderepressionofNFAT
signaling. NFAT4 (and possibly other NFATs)
arephosphorylatedbycyclinD3–CDK6 ( 115 ).
Inhibition of CDK4/6 decreased phosphoryl-
ation of NFATs, resulting in their nuclear tran-
slocation and enhanced transcriptional activity.
This caused up-regulation of NFAT targets,
resulting in T cell activation, which enhanced
the antitumor immune response. In addition,
CDK4/6 inhibitors increased the infiltration
of effector T cells into tumors, likely because
of elevated levels of chemokines CXCL9 and
CXCL10 after CDK4/6 inhibitor treatment
( 115 ). Abemaciclib treatment also induced in-
flammatory and activated T cell phenotypes in
tumors and up-regulated the expression of im-
mune checkpoint proteins CD137, PD-L1, and
TIM-3 on CD4+and CD8+cells ( 116 ).
CDK4/6 inhibition also caused up-regulation
of PD-L1 protein expression in tumor cells
( 117 , 118 ). This effect was shown to be inde-
pendent of RB1 status in the tumor. Mecha-
nistically, CDK4/6 phosphorylates and stabilizes
SPOP, which promotes PD-L1 polyubiquiti-
nation and degradation ( 118 ). Cyclin D–CDK4
also represses expression of PD-L1 through
RB1. Specifically, cyclin D–CDK4/6-mediated
phosphorylation of RB1 on S249/T252 pro-
motes binding of RB1 to NF-kB protein p65,
and this represses the expression of a subset
NF-kB–regulated genes, including PD-L1 ( 119 ).
These observations prompted tests of the
efficacy of combining CDK4/6 inhibitors with
antibodies that elicit immune checkpoint block-
ade. Indeed, treatment of mice bearing auto-
chthonous breast cancers, or cancer allografts,
with CDK4/6 inhibitors together with anti-
PD-1/PD-L1 antibodies enhanced the efficacy
of immune checkpoint blockade and led to
complete tumor regression in a high propor-
tion of animals ( 114 , 115 , 118 ). Conversely, ac-
tivation of the cyclin D–CDK4 pathway by
genomic lesions in human melanomas corre-
lated with resistance to anti–PD-1 therapy ( 117 ).
Some authors did not observe synergy when
abemaciclib was administered concurrently
with immune checkpoint inhibitors in allograft
tumor models ( 116 , 120 ). However, a strong
synergistic antitumor effect was detected when
abemaciclib was administered first (and con-
tinued) and anti–PD-L1 antibody was admin-
istered later. The combined treatment inducedimmunological memory, as mice that under-
went tumor regression were resistant to re-
challenge with the same tumor ( 116 ). Abemaciclib
plus anti–PD-L1 treatment increased infiltra-
tion of CD4+and CD8+T cells into tumors,
and increased the expression of major histo-
compatibility complex class I (MHC-I) and
MHC-II on tumor cells and on macrophages
and MHC-I on dendritic cells ( 116 ). In the case
of anti–CTLA-4 plus anti–PD-1 treatment in
melanoma allograft model, the synergistic ef-
fect was observed when immune checkpoint
inhibitor treatment was started first, followed
by abemaciclib ( 120 ).
The synergistic antitumor effect of PI3K and
CDK4/6 inhibitors in TNBC is mediated, in
part, by enhancement of tumor immunogeni-
city ( 121 ). Combined treatment of TNBC cells
with ribociclib plus the PI3K inhibitor apelisib
synergistically up-regulated the expression of
immune-related pathways in tumor cells, in-
cluding proteins involved in antigen presenta-
tion. Co-treatment of tumor-bearing mice also
decreased proliferation of CD4+FOXP3+Treg
cells, increased activation of intratumoral CD4+
and CD8+T cells, increased the frequency of
tumor-infiltrating NKT cells, and decreased the
numbers of intratumoral immunosuppressive
myeloid-derived suppressor cells. Moreover,
combined treatment strongly augmented the
response to immune checkpoint therapy with
PD-1 and CTLA-4 antibodies ( 121 ).
Single-cell RNA sequencing of human mela-
nomas identified an immune resistance prog-
ram expressed by tumor cells that correlates
with T cell exclusion from the tumor mass and
immune evasion by tumor cells. The program
can predict the response of tumors to immune
checkpoint inhibitors. Treatment of human
melanoma cells with abemaciclib repressed this
program in an RB1-dependent fashion ( 120 ).
Together, these findings indicate that CDK4/6
inhibitors may convert immunologically“cold”
tumors into“hot”ones. The most pressing issue
is to validate these findings in a clinical set-
ting. The utility of combining CDK4/6 inhibi-
tors with PD-1 or PD-L1 antibodies is currently
being evaluated in several clinical trials. Note
that the effects of CDK4/6 inhibition on the
immune system of the host are independent
of tumor cell RB1 status, raising the possibility
of using CDK4/6 inhibitors to also boost the
immune response against RB1-negative tumors.CDK4/6 inhibitors in clinical trials
Table 3 summarizes major clinical trials with
CDK4/6 inhibitors. Given early preclinical data
indicating that breast cancers—in particular,
the hormone receptor–positive ones—are very
sensitive to CDK4/6 inhibition (as discussed
above), many clinical trials have focused on
this cancer type. Most studies have evaluated
CDK4/6 inhibitors administered together
with anti-estrogens (the aromatase inhibitorsFasslet al.,Science 375 , eabc1495 (2022) 14 January 2022 7 of 19
RESEARCH | REVIEW