Articles

The cyclin-dependent kinase 4/6 inhibitor palbociclib in

combination with letrozole versus letrozole alone as
first-line treatment of oestrogen receptor-positive,
HER2-negative, advanced breast cancer
(PALOMA-1/TRIO-18): a randomised phase 2 study
Richard S Finn, John P Crown, Istvan Lang, Katalin Boer, Igor M Bondarenko, Sergey O Kulyk, Johannes Ettl, Ravindranath Patel, Tamas Pinter,
Marcus Schmidt, Yaroslav Shparyk, Anu R Thummala, Nataliya L Voytko, Camilla Fowst, Xin Huang, Sindy T Kim, Sophia Randolph, Dennis J Slamon

Summary
Background Palbociclib (PD-0332991) is an oral, small-molecule inhibitor of cyclin-dependent kinases (CDKs) 4 and 6 Lancet Oncol 2014
with preclinical evidence of growth-inhibitory activity in oestrogen receptor-positive breast cancer cells and synergy Published Online
with anti-oestrogens. We aimed to assess the safety and efficacy of palbociclib in combination with letrozole as first- December 16, 2014
http://dx.doi.org/10.1016/
line treatment of patients with advanced, oestrogen receptor-positive, HER2-negative breast cancer.
S1470-2045(14)71159-3
See Online/Comment
Methods In this open-label, randomised phase 2 study, postmenopausal women with advanced oestrogen receptor- http://dx.doi.org/10.1016/
positive and HER2-negative breast cancer who had not received any systemic treatment for their advanced disease S1470-2045(14)71188-X
were eligible to participate. Patients were enrolled in two separate cohorts that accrued sequentially: in cohort 1, See Online for podcast interview
patients were enrolled on the basis of their oestrogen receptor-positive and HER2-negative biomarker status alone, with Dennis Slamon
whereas in cohort 2 they were also required to have cancers with amplification of cyclin D1 (CCND1), loss of University of California Los
p16 (INK4A or CDKN2A), or both. In both cohorts, patients were randomly assigned 1:1 via an interactive Angeles, Los Angeles, CA, USA
(R S Finn MD,
web-based randomisation system, stratified by disease site and disease-free interval, to receive continuous oral Prof D J Slamon MD); Irish
letrozole 2·5 mg daily or continuous oral letrozole 2·5 mg daily plus oral palbociclib 125 mg, given once daily for Cooperative Oncology Research
3 weeks followed by 1 week off over 28-day cycles. The primary endpoint was investigator-assessed progression- Group, Dublin, Ireland
free survival in the intention-to-treat population. Accrual to cohort 2 was stopped after an unplanned interim (Prof J P Crown MD); Orszagos
Onkologiai Intezet, Budapest,
analysis of cohort 1 and the statistical analysis plan for the primary endpoint was amended to a combined analysis Hungary (I Lang MD); Szent
of cohorts 1 and 2 (instead of cohort 2 alone). The study is ongoing but closed to accrual; these are the results of Margit Korhaz, Onkologia,
the final analysis of progression-free survival. The study is registered with the ClinicalTrials.gov, number Budapest, Hungary
NCT00721409. (K Boer MD); Dnipropetrovsk
Medical Academy, City
Multiple-Discipline Clinical
Findings Between Dec 22, 2009, and May 12, 2012, we randomly assigned 165 patients, 84 to palbociclib plus Hospital 4, Dnipropetrovsk,
letrozole and 81 to letrozole alone. At the time of the final analysis for progression-free survival (median follow-up Ukraine
29·6 months [95% CI 27·9–36·0] for the palbociclib plus letrozole group and 27·9 months [25·5–31·1] for the (Prof I M Bondarenko MD);
Municipal Treatment and
letrozole group), 41 progression-free survival events had occurred in the palbociclib plus letrozole group and 59 in Prophylactic Institution,
the letrozole group. Median progression-free survival was 10·2 months (95% CI 5·7–12·6) for the letrozole group Donetsk, Ukraine
and 20·2 months (13·8–27·5) for the palbociclib plus letrozole group (HR 0·488, 95% CI 0·319–0·748; one-sided (S O Kulyk MD); Technical
University of Munich, Munich,
p=0·0004). In cohort 1 (n=66), median progression-free survival was 5·7 months (2·6–10·5) for the letrozole
Germany (J Ettl MD);
group and 26·1 months (11·2–not estimable) for the palbociclib plus letrozole group (HR 0·299, 0·156–0·572; Comprehensive Blood and
one-sided p<0·0001); in cohort 2 (n=99), median progression-free survival was 11·1 months (7·1–16·4) for the Cancer Center, Bakersfield, CA,
letrozole group and 18·1 months (13·1–27·5) for the palbociclib plus letrozole group (HR 0·508, 0·303–0·853; USA (R Patel MD); Petz Aladar
Megyei Oktato Korhaz, Gyor,
one-sided p=0·0046). Grade 3–4 neutropenia was reported in 45 (54%) of 83 patients in the palbociclib plus
Hungary (T Pinter MD);
letrozole group versus one (1%) of 77 patients in the letrozole group, leucopenia in 16 (19%) versus none, and University Hospital Mainz,
fatigue in four (4%) versus one (1%). Serious adverse events that occurred in more than one patient in the Mainz, Germany
palbociclib plus letrozole group were pulmonary embolism (three [4%] patients), back pain (two [2%]), and (M Schmidt MD); Lviv State
Oncologic Regional Treatment
diarrhoea (two [2%]). No cases of febrile neutropenia or neutropenia-related infections were reported during the
and Diagnostic Centre, Lviv,
study. 11 (13%) patients in the palbociclib plus letrozole group and two (2%) in the letrozole group discontinued Ukraine (Y Shparyk MD);
the study because of adverse events. Comprehensive Cancer Centers
of Nevada, Las Vegas, NV, USA
(A R Thummala MD); Kyiv City
Interpretation The addition of palbociclib to letrozole in this phase 2 study significantly improved progression-free
Clinical Oncology Centre, Kyiv,
survival in women with advanced oestrogen receptor-positive and HER2-negative breast cancer. A phase 3 trial is Ukraine (N L Voytko MD); Pfizer
currently underway. Oncology, Milan, Italy
(C Fowst MD); and Pfizer
Oncology, San Diego, CA, USA
Funding Pfizer.

www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3 1

 Articles
(X Huang PhD, S T Kim BS, Introduction
S Randolph MD) Breast cancer is a molecularly diverse disease with
Correspondence to: several defined molecular subgroups. Clinically, however,
Prof Dennis J Slamon, UCLA
Oncology, Santa Monica,
three therapeutic groups are used: those classified as
CA 90404, USA hormone receptor-positive (ie, oestrogen receptor-
dslamon@mednet.ucla.edu positive, progesterone receptor-positive, or both, with
normal HER2 expression), those classified as HER2-
positive, as defined by HER2 gene amplification or
overexpression (about 45% of these cancers can also have
variable expression of oestrogen receptors, progesterone
receptors, or both), and those classified as triple-negative
by virtue of low or absent hormone receptors and the
absence of the HER2 alteration.1 More than 1·5 million
new breast cancers are reported worldwide each year,
with roughly 60–65% of cases hormone receptor-positive,
20–25% HER2-positive, and 15–18% triple-negative.2
Hormonally directed drugs including anti-oestrogens
have been the mainstay of treatment for women with
oestrogen receptor-positive breast cancers. However,
some of these cancers have de-novo resistance to this
approach and even more will eventually develop acquired
resistance to these treatments and recur. At that point,
patients often receive chemotherapy that has little activity
in this setting and that is often associated with clinically
significant toxic effects.
New classes of molecularly targeted drugs can affect
the natural history of some subgroups of breast cancer
such as HER2-positive disease.3 However, until recently
the addition of such drugs to anti-oestrogens has not
resulted in similar improvements in hormone receptor-
positive disease. This situation changed with the approval
of everolimus for aromatase inhibitor-resistant disease,
which, when added to oestrogen blockade with
exemestane, resulted in an improvement in investigator-
assessed progression-free survival (hazard ratio [HR]
0·43, 95% CI 0·35–0·54; p<0·001) in oestrogen receptor-
positive advanced disease.4
Dysregulation of the cell cycle is one of the defined
hallmarks of cancer5 and several genetic alterations in
key cell cycle regulatory proteins have been described in
various cancers, including breast cancer.5,6 The cyclin-
dependent kinases (CDKs) are a large family of serine
threonine kinases that together with their regulatory
protein partners, the cyclins, have a crucial role in the
orderly and controlled progression through the cell cycle.
Molecular changes in several of the genes controlling the
cell cycle have been reported in various cancers, making
them an attractive potential target for new treatments.7
See Online for appendix So far, several CDK-targeted drugs have entered clinical
development, but none have shown significant activity in
solid tumours and several are associated with safety
concerns.8
Palbociclib (PD-0332991) is a reversible, oral, small-
molecule inhibitor of cyclin dependent kinases 4 and 6
(CDK4/6).9 CDK4/6 and cyclin D have a crucial role in the
regulation of the G1/S transition through regulation of the
phosphorylation state of pRb. When hyperphosphorylation
2 www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3
of pRb occurs, it causes release of transcription factors
that then allow the transition from G1 to S phase and
progression of the cell cycle.10 To investigate the therapeutic
potential for palbociclib in breast cancer, we tested its
growth inhibitory effects preclinically in a large panel of
human breast cancer cell lines and identified potent
activity in two therapeutic groups, those that were
oestrogen receptor-positive and those that were HER2-
amplified.11 This activity was associated with a major
blockade of pRb hyperphosphorylation, resulting in a G1
arrest in sensitive cells. We also noted that in combination
with the anti-oestrogen drug tamoxifen, palbociclib had
synergistic growth inhibitory activity as well as efficacy in
a model of acquired tamoxifen resistance.11 We noted
similar findings in HER2-amplified breast cancer cell
lines with trastuzumab used in combination with
palbociclib.11
Based on these data, we designed a clinical study to test
the safety and efficacy of CDK4/6 inhibition by palbociclib
in combination with anti-oestrogen drugs in oestrogen
receptor-positive breast cancer. Initially, a single-arm,
phase 1b study was done to assess the safety of palbociclib
given with letrozole in patients with oestrogen receptor-
positive, HER2-negative, advanced breast cancer and to
determine a recommended phase 2 dose of the
combination.12 The results suggested a dose and schedule
consisting of oral palbociclib 125 mg once daily for
3 weeks followed by 1 week off treatment in a 28-day
cycle, combined with the standard dose of oral letrozole
2·5 mg once daily. No drug–drug interactions were
identified and the most common treatment-related
adverse events were neutropenia, leucopenia, and
fatigue. Based on these clinical data, we planned a
randomised, open-label, phase 2 study to assess the
safety and efficacy of the palbociclib and letrozole
combination compared with letrozole alone in the first-
line treatment of women with advanced oestrogen
receptor-positive, HER2-negative breast cancer.
Methods
Study design and patients
In this international, phase 2, multicentre, open-label
randomised study (PALOMA-1/TRIO-18), post-
menopausal women (aged 18 years or older) with
oestrogen receptor-positive, HER2-negative, advanced
breast cancer were recruited from 50 sites in 12 countries
(Canada, France, Germany, Hungary, Ireland, Italy
Russia, South Africa, South Korea, Spain, Ukraine, USA;
appendix). Patients were enrolled in two separate cohorts
that accrued sequentially: in cohort 1, patients were
enrolled on the basis of their oestrogen receptor-positive
and HER2-negative biomarker status alone, whereas in
cohort 2 they were also required to have cancers with
amplification of cyclin D1 (CCND1), loss of p16 (also
known as INK4A or CDKN2A), or both. All patients were
required to have either locally recurrent disease not
amenable to surgery or evidence of metastatic disease.

Cohort 1
81 screened
15 excluded
11 did not meet inclusion criteria
4 met exclusion criteria
66 randomly assigned
34 allocated to palbociclib plus letrozole
33 received allocated intervention
1 did not receive allocated intervention
1 withdrew consent
26 discontinued intervention
8 because of adverse events
16 had objective progression or relapse
2 withdrew consent
34 analysed by intention to treat
Figure 1: Trial profile
Oestrogen receptor status was determined by routine
immunohistochemistry and HER2 status was assessed
by either fluorescent in-situ hybridisation (FISH) or
immunohistochemistry, with both eligibility markers
reported at enrolling sites. Oestrogen receptor status
was determined on the basis of either the original tissue
staining or, if available, a biopsy from the recurrence.
For enrolment into cohort 2 of the study, central
laboratory determination of CCND1 amplification or
p16 loss was required. CCND1 amplification was defined
as a CCND1-to-chromosome enumeration probe (CEP)
11 ratio greater than 1·5 and p16 loss as a p16-to-CEP9
ratio less than 0·8. These cutoffs came from an internal
analysis of British Columbia Cancer Foundation data
from 778 breast cancer cases that showed the frequency
of CCND1 amplification (more than three copies) and
p16 loss (loss of heterozygosity) to be roughly 36% in
patients with luminal B breast cancer. To establish
definitions for the genomic changes used for cohort 2,
we used a four-colour FISH assay (CCND1-to-CEP11
and p16-to-CEP9) to analyse 113 breast cancer samples
from a tumour bank and, using a cutoff of CCND1-to-
www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3
Articles
400 patients assessed for eligibility (cohorts 1 and 2)
Cohort 2
319 screened
220 excluded
203 did not meet inclusion criteria
13 met exclusion criteria
4 did not meet inclusion criteria and
met exclusion criteria
99 randomly assigned
32 allocated to letrozole 50 allocated to palbociclib plus letrozole 49 allocated to letrozole
29 received allocated intervention 50 received allocated intervention 48 received allocated intervention
3 did not receive allocated intervention 1 did not receive allocated intervention
3 withdrew consent 1 withdrew consent
28 discontinued intervention 38 discontinued intervention 41 discontinued intervention
1 because of adverse events 3 because of adverse events 1 because of adverse events
1 because of global deterioration of 5 because of global deterioration of 2 because of global deterioration of
health status health status health status
22 had objective progression or relapse 26 had objective progression or relapse 35 had objective progression or relapse
2 withdrew consent 1 died 3 withdrew consent
2 other reasons 3 withdrew consent
32 analysed by intention to treat 50 analysed by intention to treat 49 analysed by intention to treat
165 in the total intention-to-treat population
84 in the palbociclib plus letrozole group
81 in the letrozole group
CEP11 greater than 1·5 or p16-to-CEP9 less than 0·8, we
determined that 42 (37%) of 113 patients in this cohort
had CCND1 amplification, CDKN2A loss, or both.
No previous treatment for advanced disease was
permitted and all patients were required to have
measurable disease by Response Evaluation Criteria in
Solid Tumors (RECIST, version 1.0) or bone-only disease
with an Eastern Cooperative Oncology Group (ECOG)
performance status of 0 or 1 and adequate organ function
(as assessed by haematological and blood chemistry
analyses). Patients were excluded from study if they had
received letrozole as either neoadjuvant or adjuvant
treatment within the 12 months before study entry, had
received any previous treatment for advanced breast
cancer, had brain metastasis, or had previously been
treated with a CDK inhibitor.
The study was done in accordance with the
International Conference on Harmonization and Good
Clinical Practice standards. Institutional review board
approval was obtained from all participating institutions
and patients provided written informed consent before
the start of any study-specific screening procedures.
3
 Articles

Both cohorts Cohort 1 Cohort 2

Palbociclib plus Letrozole Palbociclib plus Letrozole Palbociclib plus Letrozole
letrozole (n=84) (n=81) letrozole (n=34) (n=32) letrozole (n=50) (n=49)
Median age (years) 63 (54–71) 64 (56–70) 66 (56–72) 64 (57–70) 62 (54–70) 63 (56–71)
ECOG performance status
0 46 (55%) 45 (56%) 23 (68%) 20 (63%) 23 (46%) 25 (51%)
1 38 (45%) 36 (44%) 11 (32%) 12 (38%) 27 (54%) 24 (49%)
Disease stage
III 2 (2%) 1 (1%) 2 (6%) 0 0 1 (2%)
IV 82 (98%) 80 (99%) 32 (94%) 32 (100%) 50 (100%) 48 (98%)
Disease site*
Visceral 37 (44%) 43 (53%) 10 (29%) 11 (34%) 27 (54%) 32 (65%)
Bone only 17 (20%) 12 (15%) 7 (21%) 6 (19%) 10 (20%) 6 (12%)
Other (non-visceral) 30 (36%) 26 (32%) 17 (50%) 15 (47%) 13 (26%) 11 (23%)
Disease-free interval*
>12 months from adjuvant treatment to recurrence 25 (30%) 30 (37%) 10 (29%) 10 (31%) 15 (30%) 20 (41%)
≤12 months from adjuvant treatment to recurrence 59 (70%) 51 (63%) 24 (71%) 22 (69%) 35 (70%) 29 (59%)
or de-novo advanced disease
De-novo advanced disease only 44 (52%) 37 (46%) 19 (56%) 17 (53%) 25 (50%) 20 (41%)
Previous systemic treatment
None 44 (52%) 37 (46%) 19 (56%) 17 (53%) 25 (50%) 20 (41%)
Chemotherapy 34 (40%) 37 (46%) 11 (32%) 14 (44%) 23 (46%) 23 (47%)
Hormonal 27 (32%) 28 (35%) 11 (32%) 11 (34%) 16 (32%) 17 (35%)
Tamoxifen 24 (29%) 24 (30%) 8 (24%) 8 (25%) 16 (32%) 16 (33%)
Anastrozole 8 (10%) 11 (14%) 4 (12%) 5 (16%) 4 (8%) 6 (12%)
Letrozole 2 (2%) 1 (1%) 0 0 2 (4%) 1 (2%)
Exemestane 4 (5%) 2 (2%) 3 (9%) 1 (3%) 1 (2%) 1 (2%)

Data are n (%) or median (IQR). ECOG=Eastern Cooperative Oncology Group. *Based on case report form data.

Table 1: Baseline characteristics (intention-to-treat population)

Randomisation and masking effects, study withdrawal, or death. Dose interruptions

Patients were randomly allocated 1:1 to receive either
palbociclib plus letrozole or letrozole alone. The
investigator or other member of the research staff used
an interactive web-based randomisation system to
register and randomly assign patients with
two stratifications factors: disease site (visceral, bone
only, or other) and disease-free interval (greater than
12 months from the end of adjuvant treatment to
recurrence vs 12 months or less from the end of
adjuvant treatment to recurrence or de-novo metastatic
disease). The randomisation system generated the
random assignment of the two treatments in a block
size of six for each of the stratification levels. Although
this was an open-label study, the randomisation codes
were only released at the time of interim and final
analyses and crossover was not allowed at any time.
Procedures
Patients randomly allocated to letrozole received oral
letrozole 2·5 mg once daily. Those allocated to palbociclib
plus letrozole received the same dose of letrozole plus
oral palbociclib 125 mg, given once daily for 3 weeks
followed by 1 week off in 28-day cycles. Study treatment
continued until disease progression, unacceptable toxic
and reductions were allowed for management of toxic
effects (appendix).
Tumour assessments were done locally at screening
and every 8 weeks and consisted of CT or MRI scan of
the chest, abdomen, and pelvis; radiography for bone
lesions (when applicable); and clinical assessment of
cutaneous disease. Bone scans were done at baseline and
every 12 weeks. All patients with tumour responses were
required to have response confirmation no less than
4 weeks after documentation of the initial response
report. Assessment of adverse events included incidence,
severity (graded by the National Cancer Institute
Common Terminology Criteria for Adverse Events
version 3.0), timing, seriousness, and relatedness to
study drug. Haematological and blood chemistry
analyses were done every 2 weeks for the first two
treatment cycles and at the beginning of each cycle
thereafter.
Outcomes
The primary endpoint was investigator-assessed
progression-free survival, defined as the time from
randomisation to radiological disease progression or
death on study. Secondary efficacy endpoints were

4 www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3

 Articles

objective response (by RECIST version 1.0), clinical

Both cohorts
benefit (as defined by the sum of complete plus partial 100 Palbociclib plus letrozole
responses and stable disease for 24 weeks or more), Letrozole
90
duration of response, and overall survival. Additional
secondary endpoints were safety and tissue and serum 80
biomarker analyses. Finally, we also assessed patient-

Progression-free survival (%)

70
reported outcomes using the modified Brief Pain
60
Inventory (Short Form; mBPI-sf) done on day 1 of each
treatment cycle; the mBPI-sf was used to capture whether 50
palbociclib adds to the commonly reported adverse event 40
seen with aromatase inhibitors (myalgias and joint pain).
30

Statistical analysis 20
We used a two-part study design (ie, two sequential 10
cohorts) to allow us to assess both the activity of the HR 0·488 (95% CI 0·319–0·748; one-sided p=0·0004)
0
palbociclib plus letrozole combination and to determine 0 4 8 12 16 20 24 28 32 36 40
whether further patient selection on the basis of additional Number at risk
biomarkers (CCND1 or p16) was warranted. We planned Palbociclib plus letrozole 84 67 60 47 36 28 21 13 8 5 1
Letrozole 81 48 36 28 19 14 6 3 3 1
to enrol 60 patients (30 per treatment group) into cohort 1
to provide initial safety and efficacy (progression-free Cohort 1
survival) data in patients with oestrogen receptor-positive, 100
HER2-negative, advanced breast cancer. In cohort 2, we 90
planned to include 150 patients (75 per treatment group)
80
who also had CCND1 gene amplification or loss of p16.
Progression-free survival (%)

Cohort 1 was intended to be an exploratory analysis, and 70

the analysis of the primary endpoint was initially intended 60
to be based on cohort 2 only. Assuming 114 progression-
50
free survival events in cohort 2 and using a one-sided α of
0·10, a sample size of 150 would have 80% power to detect 40
an HR of 0·67 (palbociclib plus letrozole vs letrozole 30
alone), including one futility interim analysis. This HR
20
would represent a median progression-free survival of
9 months in the control group and 13·5 months in the 10
HR 0·299 (95% CI 0·156–0·572; one-sided p<0·0001)
experimental group.13 0
However, an unplanned interim analysis of cohort 1 0 4 8 12 16 20 24 28 32 36 40
Number at risk
based on 31 progression-free survival events was done Palbociclib plus letrozole 34 26 23 18 15 13 11 8 8 5 1
when we noted that almost twice as many patients in the Letrozole 32 15 10 8 5 4 4 3 3 1
control group were coming off the study because of
Cohort 2
disease progression. This interim analysis showed 100
clinically meaningful activity of the palbociclib plus
90
letrozole combination compared with letrozole alone
(HR 0·35, 95% CI 0·170·72, p=0·006). These preliminary 80
results from cohort 1 also suggested that further patient
Progression-free survival (%)

70
selection based on CCND1 amplification or p16 loss was
60
unlikely to further improve patient outcome over the use
of oestrogen receptor and HER2 status alone (HR with 50
CCND1 or p16 copy changes 0·37 [95% CI 0·10–1·40; 40
p=0·13] vs HR with no CCND1 or p16 copy changes 0·19
30
[0·05–0·67; p=0·0045]). As a result, we stopped further
enrolment into cohort 2 and amended the statistical 20
analysis plan such that the primary endpoint would be 10
analysed in cohorts 1 and 2 combined instead of cohort 2 HR 0·508 (95% CI 0·303–0·853; one-sided p=0·0046)
0
alone. These study changes were made prospectively 0 4 8 12 16 20 24 28 32 36 40
without any efficacy results from cohort 2 and were Time (months)
Number at risk
Palbociclib plus letrozole 50 41 37 29 21 15 10 5
Figure 2: Progression-free survival (intention-to-treat population) Letrozole 49 33 26 20 14 10 2
HR=hazard ratio.

www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3 5

 Articles

Palbociclib plus Letrozole Hazard ratio Interaction

letrozole (95% CI) p value*
Patients Events Patients Events
All patients (intention-to-treat population) 84 41 81 59 0·488 (0·319–0·748)
Cohort
1 34 15 32 25 0·299 (0·156–0·572)
0·14
2 50 26 49 34 0·508 (0·303–0·853)
Age group (years)
<65 years 47 24 42 35 0·315 (0·184–0·539)
0·34
≥65 years 37 17 39 24 0·505 (0·269–0·948)
Baseline ECOG performance status
0 46 21 45 31 0·434 (0·246–0·766)
0·78
1 38 20 36 28 0·398 (0·220–0·721)
Disease site
Visceral 37 21 43 34 0·547 (0·317–0·944)
Bone Only 17 5 12 7 0·294 (0·092–0·945) 0·44
Other 30 15 26 18 0·402 (0·200–0·808)
Previous chemotherapy
Yes 34 17 37 24 0·479 (0·255–0·898)
0·75
No 50 24 44 35 0·397 (0·234–0·671)
Previous antihormonal therapy
Yes 27 12 28 19 0·460 (0·222–0·956)
0·88
No 57 29 53 40 0·397 (0·244–0·646)
Previous systemic therapy
Yes 40 20 44 28 0·539 (0·302–0·962)
0·36
No 44 21 37 31 0·341 (0·194–0·599)
Time from end of adjuvant treatment to disease recurrence
≤12 months (including de-novo presentation) 59 31 51 39 0·418 (0·259–0·674)
0·95
>12 months 25 10 30 20 0·399 (0·185–0·858)
0·34
≤12 months (excluding de-novo presentation) 15 7 14 5 0·765 (0·232–2·523)

0·062 0·125 0·250 0·500 1·000 2·000 4·000

Favours palociclib plus letrozole Favours letrozole

Figure 3: Subgroup analysis for progression-free survival

ECOG=Eastern Cooperative Oncology Group. *Two-sided p value.

overseen and approved by the study steering committee.
At the time enrolment was stopped, 165 patients had
been randomly assigned (66 patients in cohort 1 and 99
patients in cohort 2). Based on the same original
assumption that palbociclib plus letrozole would increase
progression-free survival from 9 months to 13·5 months
compared with letrozole alone, this sample size would
have 80% power to detect an HR of 0·67 based on
114 progression-free survival events in the final analysis.
A second interim analysis was added with these
protocol amendments, which was to be done when about
half of the expected number of progression-free survival
events across both cohorts (ie, about 57 of 114 total
events) had occurred. At the time of the second interim
analysis, 61 events had occurred and the HR for
progression-free survival for the entire intention-to-treat
population was 0·37 (95% CI 0·21–0·63; one-sided
p<0·0001). After these analyses were done, we noted a
substantial fall in the event rate over time and therefore
made another adjustment to the final analysis plan such
that the final analysis of progression-free survival would
be done when 95 progression-free survival events had
accumulated, giving greater than 98% power to detect an
HR of 0·50 at a one-sided α of 0·10, or 75% power to
detect an HR of 0·67.
We adjusted the significance level for the final analysis
for the interim analyses using the Lan-DeMets procedure
with an O’Brien-Fleming stopping boundary. At the final
analysis of progression-free survival, we used a gate-
keeping procedure for hypotheses testing in a hierarchical
approach to further control for family-wise errors. This
process began with assessment of all randomly assigned
patients (in cohorts 1 and 2 combined). If the null
hypothesis was rejected, then the Holm procedure would
be used to test the same hypotheses for cohorts 1 and 2 as
two separate studies. Using this approach, we compared
progression-free survival data between the treatment
groups using a stratified log-rank test with stratification
for site of disease, disease-free interval, and study cohort.
We estimated the HR using the Cox proportional hazards
regression model; the proportionality of hazards assump-
tion was verified14 and the results were satisfactory.

6 www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3

 Articles

To explore the effect of prespecified baseline prognostic
factors on progression-free survival, we did a multivariate
analysis using the Cox regression model. The primary
and secondary efficacy analyses were done in the
intention-to-treat population. The safety analyses were
done for all randomly assigned patients who received at
least one dose of the study treatment. We only controlled
the type I error for the analysis of primary endpoint, not
for any of the secondary endpoints. We did seven
prespecified sensitivity analyses for progression-free
survival (unstratified analysis; analysis stratified per case
report form data; including symptomatic deterioration as
disease progression; including disease progression or
death after 28 days of treatment discontinuation as
disease progression; forced progression-free survival
times to the planned assessment times; as-treated
population analysis; and multivariate analysis).
All statistical analyses were done with the SAS version
9.2 or later.
The study is registered with the ClinicalTrials.gov,
number NCT00721409.
Role of the funding source
The funder provided funding to the investigators for study
design, conduct, treatment administration, and data
collection. The study database was held by the funder. The
steering committee that oversaw the conduct of the study
consisted of the principal investigator (RSF), senior author
(DJS), an independent statistician, two additional investi-
gators, and three representatives from the funder (one
clinician, one operations representative, and a statistician).
The study steering committee was involved and in all
discussions about study conduct. All authors had
unrestricted access to the raw and final study data, and
Palbociclib
plus letrozole
Intention-to-treat population*
Complete response 1 (1%)
Partial response 35 (42%)
Stable disease 37 (44%)
were responsible for data interpretation, preparation of
the report, and the decision to submit for publication. The
authors attest to study completeness and the accuracy of
the data and data analysis.
Results
Between Dec 22, 2009, and May 12, 2012, 165 women
were randomly assigned, 84 to receive letrozole plus
palbociclib and 81 to receive letrozole alone (figure 1).
Baseline demographic characteristics and established
prognostic factors of the intention-to-treat population
were generally well balanced, although there were slight
imbalances in disease site, disease-free interval, and
previous treatment (table 1). About half of the patients in
each group had never received either adjuvant or
neoadjuvant systemic treatment. Conversely, a third of
patients in each group had received previous anti-
hormonal treatment, with half of these individuals
having previously received aromatase inhibitors.
As of the data cutoff for the final analysis (Nov 29,
2013), median follow-up was 29·6 months (95% CI
27·9–36·0) for the palbociclib plus letrozole group and
27·9 months (25·5–31·1) for the letrozole group, with
19 (23%) of 84 patients in the palbociclib plus letrozole
group and eight (10%) of 81 in the letrozole group
remaining on treatment. At the time of the final analysis
for progression-free survival, 41 progression-free survival
events had occurred in the palbociclib plus letrozole
group and 59 in the letrozole group. Median progression-
free survival was 20·2 months (95% CI 13·8–27·5) for
the palbociclib plus letrozole group and 10·2 months
(5·7–12·6) for the letrozole alone group (HR 0·488,
95% CI 0·319–0·748; one-sided p=0·0004; figure 2). For
patients in cohort 1, median progression-free survival
was 26·1 months (95% CI 11·2–not estimable [NE]) for
the combination and 5·7 months (95% CI 2·6–10·5) for
Letrozole
100 Palbociclib plus letrozole
Letrozole
90
1 (1%)
26 (32%) 80
30 (37%) 70
Overall survival (%)

Stable disease ≥24 weeks 32 (38%) 20 (25%)

60
Stable disease <24 weeks 5 (6%) 10 (12%)
50
Progressive disease 3 (4%) 18 (22%)
Indeterminate 8 (10%) 6 (7%) 40
Patients with measurable disease† 30
Complete response 1 (2%) 0
20
Partial response 35 (54%) 26 (39%)
Stable disease 20 (31%) 22 (33%) 10
HR 0·813 (95% CI 0·492–1·345; two-sided p=0·42)
Progressive disease 2 (3%) 15 (23%) 0
0 4 8 12 16 20 24 28 32 36 40 44
Indeterminate 7 (11%) 3 (5%)
Time (months)
Data are n (%). *n=84 in the palbociclib plus letrozole group, n=81 in the letrozole Number at risk
Palbociclib plus letrozole 84 80 78 73 68 65 47 35 22 17 7 2
alone group. †n=65 in the palbociclib plus letrozole group, n=66 in the letrozole
Letrozole 81 76 74 67 64 59 37 23 14 12 5 1
alone group.

Table 2: Best overall response Figure 4: Overall survival (intention-to-treat population)

HR=hazard ratio.

www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3 7

 Articles

letrozole alone (HR 0·299, 95% CI 0·156–0·572; one- (7·1–16·4) for letrozole alone (HR 0·508, 95% CI
sided p<0·0001; figure 2). For patients in cohort 2, 0·303–0·853; one-sided p=0·0046; figure 2). The effect
median progression-free survival was 18·1 months (95% of the combination treatment relative to letrozole alone
CI 13·1–27·5) for the combination and 11·1 months was consistent across all demographic subgroups and
patient baseline prognostic factors, apart from patients
with disease recurrence 12 months or less from the end
Palbociclib plus letrozole (n=83) Letrozole (n=77)
of adjuvant treatment, although this subgroup is limited
Grade 1–2 Grade 3 Grade 4 Grade 1–2 Grade 3 Grade 4 by small numbers in both groups (figure 3). The results
Any adverse event 19 (23%) 49 (59%) 14 (17%) 49 (64%) 16 (21%) 0 from prespecified sensitivity analyses were consistent
Neutropenia 17 (20%) 40 (48%) 5 (6%) 3 (4%) 1 (1%) 0 with those of the main analysis (data not shown).
Leucopenia 20 (24%) 16 (19%) 0 2 (3%) 0 0 Table 2 shows best responses to treatment. A greater
Fatigue 30 (36%) 2 (2%) 2 (2%) 17 (22%) 1 (1%) 0 proportion of patients in the palbociclib plus letrozole
Anaemia 24 (29%) 4 (5%) 1 (1%) 4 (5%) 1 (1%) 0 group than in the letrozole group had an objective
Nausea 19 (23%) 2 (2%) 0 9 (12%) 1 (1%) 0 response to treatment, both in the intention-to-treat
Arthralgia 18 (22%) 1 (1%) 0 10 (13%) 2 (3%) 0 population (36 [43%, 95% CI 32–54] vs 27 [33%, 23–45];
Alopecia 18 (22%) NA NA 2 (3%) NA NA p=0·13) and in the population with measurable disease
Diarrhoea 14 (17%) 3 (4%) 0 8 (10%) 0 0 (36 [55%, 43–68] vs 26 [39%, 28–52]; one-sided p=0·047).
Hot flush 17 (21%) 0 NA 9 (12%) 0 NA Similarly, a greater proportion of patients in the
Thrombocytopenia 12 (14%) 2 (2%) 0 1 (1%) 0 0 intention-to-treat population achieved clinical benefit
Decreased appetite 12 (14%) 1 (1%) 0 5 (6%) 0 0
(68 [81%, 95% CI 71–89] vs 47 [58%, 47–69]; one-sided
Dyspnoea 11 (13%) 2 (2%) 0 5 (6%) 1 (1%) 0
p=0·0009). The median duration of response for patients
Nasopharyngitis 13 (16%) 0 0 8 (10%) 0 0
who had a complete or partial response was 20·3 months
Back pain 11 (13%) 0 1 (1%) 11 (14%) 1 (1%) 0
(95% CI 13·4–25·8) for the palbociclib plus letrozole
group and 11·1 months (9·3–31·6) for the letrozole
Headache 12 (14%) 0 0 8 (10%) 0 0
group. At the same time as the final progression-free
Vomiting 12 (14%) 0 0 2 (3%) 1 (1%) 0
survival analysis, we also assessed overall survival.
Asthenia 9 (11%) 2 (2%) 0 3 (4%) 0 0
Median overall survival was 37·5 months (95% CI
Bone pain 8 (10%) 1 (1%) 1 (1%) 3 (4%) 0 0
28·4–NE; 30 events) in the palbociclib plus letrozole
Constipation 10 (12%) 0 0 7 (9%) 0 0
group and 33·3 months (26·4–NE; 31 events) in the
Cough 10 (12%) 0 0 8 (10%) 0 0
letrozole alone group (HR 0·813, 95% CI 0·492–1·345;
Stomatitis 10 (12%) 0 0 2 (3%) 0 0
two-sided p=0·42; figure 4).
Epistaxis 9 (11%) 0 0 1 (1%) 0 0
The most common adverse events reported for the
Influenza 8 (10%) 1 (1%) 0 1 (1%) 0 0
palbociclib plus letrozole group were neutropenia,
Musculoskeletal pain 8 (10%) 1 (1%) 0 5 (6%) 0 0
leucopenia, and fatigue (table 3). All 83 patients who
Upper respiratory tract infection 8 (10%) 1 (1%) 0 2 (3%) 0 0 received palbociclib plus letrozole had at least one
Dizziness 8 (10%) 0 0 3 (4%) 0 0 adverse event, compared with 65 (84%) of 77 who received
Peripheral neuropathy 8 (10%) 0 0 4 (5%) 0 0 letrozole alone. Despite the increase in all grades of
Oropharyngeal pain 8 (10%) 0 0 1 (1%) 0 0 neutropenia and leucopenia with palbociclib plus
Pain in extremity 8 (10%) 0 0 6 (8%) 0 0 letrozole, no cases of neutropenic fever were reported.
Data are n (%) unless otherwise specified. One (1%) grade 5 adverse event occurred in the palbociclib plus letrozole group Other adverse events (of any cause) that were increased
(from disease progression) and none in the letrozole group. Additional grade 4 adverse events reported in the palbociclib in the palbociclib plus letrozole group included anaemia,
plus letrozole group that are not shown in the table are neutropenia (five [6%]), pulmonary embolism (four [5%]), bone nausea, arthralgia, and alopecia, but most of these were
pain (one [1%]), chest pain (one [1%]), increased blood uric acid increased (1 [1%]), gangrene (one [1%]), and humerus
fracture (one [1%]). No additional grade 4 adverse events were reported in the letrozole group. Additional grade 3
low grade (table 3). Of these adverse events, the difference
adverse events reported in the palbociclib plus letrozole group that are not shown in the table are abdominal pain (two between treatment groups was significant only for
[2%]), increased γ-glutamyl transferase (two [2%]), decreased white blood cell count (two [2%]), increased aspartate anaemia (two-sided p<0·0001) and alopecia (two-sided
aminotransferase (one [1%]), pain (one [1%]), spinal pain (one [1%]), increase blood lactate dehydrogenase (one [1%]),
p=0·0002). Serious adverse events that occurred in more
joint stiffness (one [1%]), pneumonia (one [1%], wound (one [1%]), decreased blood potassium (one [1%]), cancer pain
(one [1%]), ischaemic colitis (one [1%]), coronary artery disease (one [1%]), exertional dyspnoea (one [1%]), fallopian than one patient in the palbociclib plus letrozole group
tube cancer (one [1%]), fractured sacrum (one [1%]), gastrointestinal disorder (one [1%]), granulocytopenia (one [1%]), were pulmonary embolism (three [4%] patients), back
hypermagnesaemia (one [1%]), intervertebral disc protrusion (one [1%]), nephrolithiasis (one [1%]), decreased pain (two [2%]), and diarrhoea (two [2%]). No serious
neutrophil count (one [1%]), pathological fracture (one [1%]), periostitis (one [1%]), petechiae (one [1%]), renal disorder
(one [1%]), staphylococcal bacteraemia (one [1%]), and urethral obstruction (one [1%]). Additional grade 3 adverse
adverse events occurred in more than one patient in the
events reported in the letrozole group that are not shown in the table are pleural effusion (two [3%]), hypertension (one letrozole group.
[1%]), chest pain (one [1%]), decreased haemoglobin (one [1%]), pain (one [1%]), spinal pain (one [1%]), dysphagia (one 27 (33%) patients in the palbociclib plus letrozole had
[1%]), hypocalcaemia (one [1%]), mucosal inflammation (one [1%]), neck pain (one [1%]), pelvic pain (one [1%]), cardiac
dose interruptions because of adverse events, compared
failure (one [1%]), hip fracture (one [1%]), ileus (one [1%]), laceration (one [1%]), oesophageal achalasia (one [1%]),
pneumothorax (one [1%]), and subcutaneous emphysema (one [1%]). with only three (4%) patients in the letrozole group. In
the combination group, 37 (45%) patients required a
Table 3: Most common all-cause adverse events that occurred in at least 10% of patients (safety
delay in the start of a subsequent treatment cycle because
population)
of an adverse event and 33 (40%) patients had a dose

8 www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3


reduction. However, the mean relative dose intensity for
palbociclib in the combination group was 94% (SD 26).
Cycle delays and dose reductions are not applicable to the
letrozole group.
The main reason for study discontinuation in both
treatment groups was disease progression (42 [50%]
patients in the palbociclib plus letrozole group vs 57 [70%]
patients in the letrozole group). 11 (13%) patients in the
palbociclib plus letrozole group and two (2%) patients in
the letrozole group discontinued the study because of an
adverse event. Of these discontinuations, six (7%)
patients in palbociclib plus letrozole group and two (2%)
patients in the letrozole group discontinued because of
treatment-related adverse events. One death occurred
during the study in the palbociclib plus letrozole group
because of disease progression; no treatment-related
deaths occurred. We noted no significant differences in
pain severity or the effect of pain on daily activities
between the two treatment groups.
Discussion
The results of this open-label, phase 2 study show that
patients with oestrogen receptor-positive, HER2-negative
advanced breast cancer had significantly longer
progression-free survival when treated with palbociclib
and letrozole than when treated with letrozole alone.
Additionally, the proportions of patients with an objective
response and clinical benefit were greater in the
combination group than in the letrozole alone group.
The study was not powered to detect an overall survival
advantage and few overall survival events had occurred at
the time of this analysis; however, the initial data suggest
no detrimental effect on overall survival with the addition
of palbociclib in the first-line setting (panel).
Hormone directed drugs have been the mainstay of
treatment for advanced oestrogen receptor-positive
breast cancer for more than four decades. Improvements
in clinical outcomes have occurred with several drugs
that target either specific hormone production (ie,
ligands or the hormone receptor pathway), including
tamoxifen, steroidal and non-steroidal aromatase
inhibitors, and fulvestrant.15 Despite efforts to further
improve clinical outcomes for patients with oestrogen
receptor-positive breast cancer with drugs that target
other pathways thought to have a role in the development
of resistance to hormone drugs, most results have been
largely disappointing, including efforts to target the
HER1 and HER2 pathways, angiogenesis, and IGFR.16–18
Recently, however, targeting of mTOR, a crucial
component of the PI3K pathway, with everolimus, used
in combination with a steroidal aromatase inhibitor,
resulted in improved progression-free survival, although
not overall survival, in patients with oestrogen receptor-
positive, advanced disease who had progressed on anti-
hormonal treatment.4,19
Our findings need to be interpreted in the context of
the limitations of the study design. Specifically, the study
www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3
Articles
is open-label and did not use central radiology review to
prospectively assess the primary endpoint, but rather a
retrospective, masked, independent review after accrual
was completed. This analysis was limited by the fact that
scans were obtained retrospectively and were not used to
make on-treatment decisions. On-treatment decisions
were made on the basis of scan reviews at the individual
study sites. We noted some baseline imbalances based
on the case report form data; however, sensitivity analyses
including multivariate analysis to control for baseline
factors consistently showed treatment benefit in the
combination group across all demographic and clinical
subgroups.
These clinical results are supported by preclinical data11
that provide a clear biological rationale for the
development of palbociclib in this patient population. As
with most other molecularly targeted drugs in oncology,
the greatest gains are often seen when rational and
appropriate patient selection can be used prospectively.
Preclinical data11 with palbociclib identified oestrogen
receptor-positive breast cancer cells as one of the subtypes
most sensitive to CDK4/6 inhibition, the other being
HER2 amplification. Our results provide clinical
validation of these preclinical data and support the further
development of palbociclib for the oestrogen receptor-
positive, HER2-negative subgroup of breast cancers.
Genetic changes in cyclin D1 and p16 are known to
occur in breast cancer and might have a role in the
further selection of patients for treatment with a CDK4/6
inhibitor. In cohort 2, we investigated the potential for
these genetic changes to be used to improve patient
selection beyond use of oestrogen receptor-positive status
alone. However, our results did not substantiate this
hypothesis. This analysis confirmed that oestrogen
receptor positivity is currently the best and most effective
predictive marker for the identification of patients likely
to respond to CDK4/6 inhibition. Further biomarker
Panel: Research in context
Systematic review
We searched PubMed for reports of clinical trials published in English between Nov 1,
2004, and Nov 1, 2014, using the terms “breast cancer” and “cyclin-dependent kinase
inhibitors”. We did not identify any studies that investigated cyclin-dependent kinase
(CDK) inhibitors specifically in women with breast cancer, despite laboratory data
suggesting a role for CDK inhibition in breast cancer.7,8
Interpretation
Our results show that palbociclib, a novel, oral, inhibitor of CDK4/6, used in combination
with letrozole resulted in longer progression-free survival than letrozole alone in
postmenopausal women with advanced oestrogen receptor-positive, HER2-negative
breast cancer. These results provide the first clinical data to show a role for CDK4/6
inhibition in breast cancer. The safety profile of the combination was predictable and
manageable. Overall, our results suggest that palbociclib in combination with letrozole
could become an important treatment option, substantially improving upon letrozole
alone for the treatment of postmenopausal women with oestrogen receptor-positive,
HER2-negative, advanced breast cancer.
9
 Articles
research designed to improve on oestrogen receptor-
positive status as the selection biomarker is ongoing;
however, in view of the large proportion of patients in
our study who achieved a clinical benefit response (more
than 80%), the benefit of additional biomarkers could be
difficult to ascertain.
Negative-selection biomarkers of resistance might be
more easily identified and will also be assessed in ongoing
and future molecular studies. One of the most important
markers of sensitivity to palbociclib is the presence of an
intact Rb pathway; however, since pRb loss is uncommon
in oestrogen receptor-positive, HER2-negative breast
cancers, it was not used as a prospective independent
biomarker for patient selection in the present study.
Potentially, relative amounts of pRb (rather than its
presence or absence) in the various breast cancer subtypes
might be of predictive importance and early preclinical
data11 suggest that this possibility should be investigated.
When comparing median progression-free survival in
the letrozole alone groups, we noted a difference between
cohorts 1 and 2, suggesting a potential predictive value
for cyclin D1 gains, p16 loss, or both in determining
response to letrozole alone. However, this finding could
simply be an artifact of the sample size in cohort 1, so
further study is necessary.
Since the initiation of this study, additional laboratory
findings have linked CDK 4/6 inhibition to endocrine
sensitivity in oestrogen receptor-positive breast cancer.20,21
The Cancer Genome Atlas has been used to identify
common changes in the Rb pathway in all breast cancer
subtypes, including the luminal oestrogen receptor-
positive, HER2-negative subgroup. However, our findings
suggest that limiting patient selection to those with
defined genetic changes in the Rb pathway might exclude
a much larger group of patients that could benefit from
CDK4/6 inhibition. Additionally, the results of a recent
phase 2, single-arm study22 of palbociclib in patients with
heavily pretreated advanced breast cancer showed single-
agent activity in some patients with oestrogen receptor-
positive, HER2-negative breast cancers.
Our results also provide useful data for the safety
profile of the combination of palbociclib and letrozole,
suggesting that adverse events are predictable and
manageable. Neutropenia, although common, was not
accompanied by serious clinical outcomes and is likely to
be the result of an on-target effect of CDK4/6 inhibition
on marrow progenitor cells. The absence of serious
complications resulting from palbociclib-associated
neutropenia probably reflects a cytostatic rather than
cytotoxic effect of the drug on bone marrow progenitor
cells,23 different from what is seen with typical cytotoxic
drugs. Additional analyses of the effect of palbociclib on
quality-of-life measures are ongoing in the context of
phase 3 studies.
Taken together, the data from this study provides a
proof of concept for the activity and safety of CDK4/6
inhibition in advanced, oestrogen receptor-positive,
10 www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3
HER2-negative breast cancer. The improvement in
progression-free survival is substantial in this population
and is accompanied by manageable toxic effects. These
data clearly warrant further investigation of the efficacy
and safety of palbociclib in combination with hormonal
blockade, both in patients with this subtype of breast
cancer and in other cancer settings. A phase 3, double-
blind, placebo-controlled study (NCT01740427) in a
similar patient population (n=650) with the aim of
confirming the present phase 2 findings is now fully
enrolled and ongoing. Additionally, other phase 3 studies
of palbociclib in combination with various anti-hormonal
drugs in additional breast cancer settings are now
ongoing (NCT01942135 and NCT01864746).
Contributors
RSF, SOK, CF, XH, STK, SR, and DJS contributed to the design of the
study. RSF, DJS, and JPC were members of the steering committee that
oversaw the conduct of the study. RSF, JPC, IMB, SOK, RP, ART, and
NLV contributed substantially to patient recruitment. RSF, JPC, IL, KB,
IMB, SOK, JE, RP, TP, MS, YS, NLV, DJS, ART, and STK contributed to
data collection. JPC, RP, and SR contributed to data review. RSF, JPC,
IMB, SOK, RP, CF, XH, STK, SR, and DJS contributed to data analysis.
RSF, IL, KB, IMB, SOK, JE, RP, TP, MS, YS, CF, XH, STK, SR, and DJS
contributed to data interpretation. RSF, JE, MS, CF, XH, STK, SR, and
DJS contributed to the writing and revision of the report. DJS searched
the scientific literature. DJS and XH contributed to production of the
figures. All authors contributed to review of the report and approved the
final submitted version.
Declaration of interests
RSF and DJS received research support from Pfizer and have served as
consultants to the same company. DJS has common stock in Pfizer.
JPC has received conference support from Pfizer. JE has received
honoraria from Pfizer for attending advisory board meetings. MS has
received research support, served as a consultant, and received honoraria
from Pfizer. He has also received personal fees from Novartis,
AstraZeneca, Roche, GlaxoSmithKline, Sanofi-Aventis, Sividon, Pierre-
Fabre, Celgene, and Eisai. CF, XH, STK, and SR are employees of Pfizer.
KB, SOK, TP, IL, IMB, YS, RP, ART, and NLV declare no competing
interests.
Acknowledgments
The study was funded by Pfizer. DJS has received funding for the
preclinical work on cyclin-dependent kinase 4/6 inhibition that
generated translational concepts for this clinical study from the US
Department of Defense Breast Cancer Innovator Award, the Revlon/
UCLA Women’s Cancer Research Program, and the Peter and Denise
Wittich Breast Cancer Program. We thank all the study participants and
their families, as well as the staff of Translational Research in Oncology
for their support. We also thank Nicholas J Robert (Virginia Cancer
Specialists, US Oncology, Fairfax, VA, USA) and Marc Buyse (IDDI,
Louvain-la-Neuve, Belgium) for their contributions. Karen Miller and
Cynthia Gobbel from Complete Healthcare Communications (Chadds
Ford, PA, USA) provided administrative support for the report
submission (funded by Pfizer).
References
1 Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human
breast tumours. Nature 2000; 406: 747–52.
2 Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global
cancer statistics. CA Cancer J Clin 2011; 61: 69–90.
3 Slamon D, Leyland-Jones B, Shak S, et al. Use of chemotherapy
plus a monoclonal antibody against HER2 for metastatic breast
cancer that overexpresses HER2. N Engl J Med 2001; 344: 783–92.
4 Baselga J, Campone M, Piccart M, et al. Everolimus in
postmenopausal hormone-receptor-positive advanced breast cancer.
N Engl J Med 2012; 366: 520–29.
5 Hanahan D, Weinberg RA. Hallmarks of cancer: the next
generation. Cell 2011; 144: 646–74.

6 Cancer Genome Atlas Network. Comprehensive molecular portraits
of human breast tumours. Nature 2012; 490: 61–70.
7 Malumbres M, Barbacid M. Cell cycle, CDKs and cancer: a
changing paradigm. Nat Rev Cancer 2009; 9: 153–66.
8 Stone A, Sutherland RL, Musgrove EA. Inhibitors of cell cycle
kinases: recent advances and future prospects as cancer
therapeutics. Crit Rev Oncog 2012; 17: 175–98.
9 Fry DW, Harvey PJ, Keller PR, et al. Specific inhibition of
cyclin-dependent kinase 4/6 by PD 0332991 and associated
antitumor activity in human tumor xenografts. Mol Cancer Ther
2004; 3: 1427–38.
10 Choi YJ, Anders L. Signaling through cyclin D-dependent kinases.
Oncogene 2014; 33: 1890–903.
11 Finn RS, Dering J, Conklin D, et al. PD 0332991, a selective cyclin D
kinase 4/6 inhibitor, preferentially inhibits proliferation of luminal
estrogen receptor-positive human breast cancer cell lines in vitro.
Breast Cancer Res 2009; 11: R77.
12 Finn R, Hurvitz S, Allison M, et al. Phase I study of PD 0332991, a
novel, oral, cyclin-D kinase (CDK) 4/6 inhibitor in combination with
letrozole, for first-line treatment of metastatic post-menopausal,
estrogen receptor-positive (ER+), human epidermal growth factor
receptor 2 (HER2)-negative breast cancer. Cancer Res 2009;
69 (24 suppl): abstr 5069.
13 Mouridsen H, Gershanovich M, Sun Y, et al. Phase III study of
letrozole versus tamoxifen as first-line therapy of advanced breast
cancer in postmenopausal women: analysis of survival and update
of efficacy from the International Letrozole Breast Cancer Group.
J Clin Oncol 2003; 21: 2101–09.
14 Grambsch PM, Therneau TM. Proportional hazards tests and
diagnostics based on weighted residuals. Biometrika 1994; 81: 515–26.
15 Chlebowski RT. Changing concepts of hormone receptor-positive
advanced breast cancer therapy. Clin Breast Cancer 2013; 13: 159–66.
www.thelancet.com/oncology Published online December 16, 2014 http://dx.doi.org/10.1016/S1470-2045(14)71159-3
Articles
16 Johnston S, Pippen J Jr, Pivot X, et al. Lapatinib combined with
letrozole versus letrozole and placebo as first-line therapy for
postmenopausal hormone receptor-positive metastatic breast
cancer. J Clin Oncol 2009; 27: 5538–46.
17 Martin M, Loibl S, von Minckwitz G, et al. Phase III trial evaluating
the addition of bevacizumab to endocrine therapy as first-line
treatment for advanced breast cancer—first efficacy results from the
LEA study. Cancer Res 2012; 72: abstr S1-7.
18 Robertson JF, Ferrero JM, Bourgeois H, et al. Ganitumab with
either exemestane or fulvestrant for postmenopausal women with
advanced, hormone-receptor-positive breast cancer: a randomised,
controlled, double-blind, phase 2 trial. Lancet Oncol 2013;
14: 228–35.
19 Piccart M, Hortobagyi GN, Campone M, et al. Everolimus plus
exemestane for hormone receptor-positive (HR+), human
epidermal growth factor receptor-2-negative (HER2–) advanced
breast cancer (BC): overall survival results from BOLERO-2.
Eur J Cancer 2014; 50 (suppl 3): S1.
20 Dean JL, McClendon AK, Hickey TE, et al. Therapeutic response to
CDK4/6 inhibition in breast cancer defined by ex vivo analyses of
human tumors. Cell Cycle 2012; 11: 2756–61.
21 Miller TW, Balko JM, Fox EM, et al. ERα-dependent E2F
transcription can mediate resistance to estrogen deprivation in
human breast cancer. Cancer Discov 2011; 1: 338–51.
22 DeMichele A, Clark AS, Heitjan D, et al. A phase II trial of an oral
CDK 4/6 inhibitor, PD0332991, in advanced breast cancer. 2013
ASCO Annual Meeting; Chicago, IL, USA; May 31–June 4, 2013.
Abstr 519.
23 Roberts PJ, Bisi JE, Strum JC, et al. Multiple roles of cyclin-
dependent kinase 4/6 inhibitors in cancer therapy. J Natl Cancer Inst
2012; 104: 476–87.
11