University of Groningen Imaging hormone receptors in metastatic breast cancer patients Venema, Clasina Marieke

(1)

Imaging hormone receptors in metastatic breast cancer patients

Venema, Clasina Marieke

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2018

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Venema, C. M. (2018). Imaging hormone receptors in metastatic breast cancer patients. Rijksuniversiteit Groningen.

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

517613-L-sub01-bw-Venema 517613-L-sub01-bw-Venema 517613-L-sub01-bw-Venema 517613-L-sub01-bw-Venema Processed on: 7-3-2018 Processed on: 7-3-2018 Processed on: 7-3-2018

Processed on: 7-3-2018 PDF page: 71PDF page: 71PDF page: 71PDF page: 71

Chapter 5

Early identification of metastatic breast cancer patients who benefit

from palbociclib combined with letrozole by means of molecular

imaging

C.M. Venema1, E.F.J. de Vries2, A.W.J.M. Glaudemans2, C. Zielstra-Dorbitz3, G.A.P. Hospers1, C.P. Schröder1

1

: Department of Medical Oncology, University of Groningen, University

Medical Center Groningen, the Netherlands 2

: Department of Nuclear Medicine and Molecular Imaging, University of

Groningen, University Medical Center Groningen, the Netherlands 3

: Department of Radiology, University of Groningen, University Medical

(3)

72

ABSTRACT

INTRODUCTION: Palbociclib plus letrozole has improved both the

progression free survival and overall response rate in metastatic breast cancer (MBC) patients. Estrogen receptor (ER) expression is the best biomarker for response to both palbociclib and letrozole. Positron emission tomography (PET) with 16α-[18F]Fluoro-17β-estradiol (FES) allows whole body assessment of ER expression, and provides insight in the heterogeneity in ER expression between lesions. We hypothesized that lesions with low FES uptake are unlikely to respond to letrozole plus palbociclib. METHODS: Post-menopausal women with ER positive MBC were eligible for this pilot study. Staging was performed by 18 F-fluorodeoxyglucose PET and contrast enhanced CT (FDG-PET/CT). In addition, a FES-PET was performed at baseline. After 8 weeks of treatment, FDG-PET/CT was repeated for response evaluation. The primary endpoint was the relation between standardized uptake value (SUV) on FES-PET, to response per lesion: on CT according to RECIST 1.1 (in case of measurable disease), or metabolic response FDG-PET according to PERCIST (in case of non-measurable disease). RESULTS: 15 patients were included of which 14 were evaluable for the primary endpoint. A total of 279 lesions were detected on conventional imaging of which 50 showed low uptake (SUV<1.5) on FES-PET. The majority of these FES negative lesions (30/50, 60%) showed progression after 8 weeks of treatment. In contrast, only 12% of FES positive lesions (27/229) showed progression after 8 weeks (p < 0.01). None of the FES negative visceral lesions showed response (negative predictive value 100%). CONCLUSION: This feasibility trial indicates that low FES uptake on FES PET in MBC lesions is related to lack of response and early progression on letrozole and palbociclib treatment, particularly in visceral lesions. FES-PET may therefore be a potential biomarker for whole body response for this combination treatment.

(4)

73

INTRODUCTION

Knowledge of the expression levels of different receptors in tumors of breast cancer patients is important for prognostic and therapeutic purposes. For example, patients with a tumor that expresses the estrogen receptor (ER) have a more favorable prognosis, and are more likely to respond to antihormonal therapy (1,2). Letrozole in combination with palbociclib, a cyclin dependent kinase 4/6 inhibitor, is considered standard treatment in hormone positive, metastatic breast cancer. This is based on improved progression free survival and overall response rates (3,4). Response rates are up to 80% when used as first line treatment. However, a phase II study showed that overall survival was not significantly improved for the combination of palbociclib plus letrozole (37.5 months) versus letrozole alone (33.3 months) (5). Furthermore, this combination therapy has some disadvantages. Forty percent of the patients experience toxicity, mostly bone-marrow depletion and fatigue. Besides toxicity to patients, there is also a financial burden for the community, as treatment is available at high costs. Recently this has led to the decline of reimbursement of the combination of palbociclib with an aromatase inhibitor in the United Kingdom by the National Institute for Health and Care Excellence (6). Biomarkers to select patients who will or will not benefit from the combination therapy are therefore urgently needed.

Currently, the only biomarker proposed for prediction of treatment response is ER expression of the tumor (3,4). Standard histological staining of ER on the primary tumor is part of standard clinical care, however in the setting of metastatic breast cancer several difficulties arise. As a result of disease heterogeneity, ER status of one (biopsied) lesion may not adequately reflect the ER status of all lesions throughout the body (7). Standard imaging modalities such as computed tomography (CT) or magnetic resonance imaging (MRI) cannot dichotomize between ER positive and ER negative lesions, but whole body positron emission tomography (PET) of ER expression with 17-[18F]fluoroestradiol (FES) is able to differentiate between ER positive and ER negative lesions and thus may be able to predict treatment response (9).

(5)

74

Therefore we hypothesized that lesions with low uptake on FES-PET will not respond to the combination of letrozole plus palbociclib. In this feasibility study, we investigated on a lesion by lesion basis whether non-invasive in vivo imaging of ER expression in metastatic breast cancer patients by means of FES-PET can predict treatment response to palbociclib plus letrozole. In addition, we explored whether the addition of FDG PET could further refine response prediction, by means of the ratio of FES to FDG uptake to identify more indolent and more aggressive tumors (Kurland), and by means of a FES heterogeneity score (Gennari)(10-12).

Furthermore we explored whether estrogen receptor mutations in blood are related to FES uptake. To date most is known about the activating ER mutations ESR1 D538G, Y537S, L536Q, Y537C, S463P or E380Q. These mutations occur in the ligand binding domain and therefore could also alter the binding capacity of FES. The presence of these mutations is related to early treatment failure and worse overall survival compared to tumors expressing wild type ER (13,14). It is unclear if these mutations result in altered FES signal. We therefore explored the relation between FES uptake in the tumor and the activating ER mutations Y537C and D538G found in circulating tumor DNA.

METHODS

This investigator-initiated study was conducted at the University Medical Center Groningen, the Netherlands, in accordance with the principles of Good Clinical Practice and the Declaration of Helsinki. The Institutional Review Board approved the protocol, and patients provided written informed consent before participation. The study was registered in clinicaltrials.gov (NCT02806050).

Study Population

Post-menopausal women with primary ER-positive, metastatic breast cancer and eligible for letrozole and palbociclib therapy were included in the study. Inclusion criteria included adequate bone marrow and organ function (absolute neutrophil count >1.0 x 109_{/L, white blood cell count >3 x}

109/L, platelet count >100 x 109/L, liver function tests <5 x upper limit of normal in case of known liver metastases, and creatinine clearance > 30

(6)

75

mL/min) and an Eastern Cooperative Oncology Group performance score of 0-2 (15). Patients were excluded when there was evidence of central nervous system metastases, prior use of CDK 4/6 inhibitors, presence of life-threatening visceral metastases, a life expectancy of less than 3 months, an active cardiac disease or a concurrent malignancy. Patients did not take any estrogen receptor ligands (i.e. fulvestrant, tamoxifen or estradiol) at least 6 weeks prior to study entry to avoid false negative scans due to the occupation or downregulation of the ER (9, 16).

Study Design

In this feasibility study all patients were staged with an FDG-PET and contrast enhanced CT of the thorax and abdomen at baseline. FES-PET was executed within 7 days prior to the start of treatment. All patients were treated with letrozole 2.5 mg daily and palbociclib 150 mg daily for 21 consecutive days followed by 7 days off treatment. After 8 weeks, response measurements with CT according to the Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1) were performed (8). For non-measurable bone lesions, also an FDG PET scan was performed and assessed according to the PET Response Criteria in Solid Tumors (PERCIST) at 8 weeks (Figure 1) (17).

(7)

76

Tumor Histology

Patients with positive ER expression in a previously resected or biopsied (primary or recurrent) tumor lesion were recruited for the study. A new metastasis biopsy was performed in patients at study entry, if feasible, except for those patients that had already undergone biopsy within 6 months prior to baseline FES-PET. Immunohistochemistry for ER was performed on stored paraffin-embedded tumor samples. ER (CONFIRM anti-Estrogen Receptor (SP1) Rabbit Monoclonal Primary Antibody, Ventana, Illkirch, France) were stained with a Ventana Benchmark automated stainer (Ventana, Illkirch, France). ER was scored according to the American Society of Clinical Oncology pathologists guideline (18).

Imaging

A whole body FDG PET/CT scan was performed at baseline and repeated after 8 weeks. Patients were required to fast for at least 6 hours prior to the scan and to have blood glucose levels < 1.2 g/l before FDG injection. The injected 18F-FDG activity was 3 MBq/kg according to EANM guidelines. Whole body (head to mid-thigh) PET/CT was performed 60 minutes after tracer injection using a Siemens 40 or 64-slice mCT (Siemens Medical Systems, Knoxville, TN) with 2 mm reconstructed spatial resolution and emission acquisition time of 3 minutes per bed position. Low dose CT was acquired for attenuation and scatter correction. Reconstructions of the scan and quantification of tracer uptake were performed according to the European association of nuclear medicine research limited (EARL) criteria for quantification measurements (19).

Whole body FES PET/CT was performed in the same manner as the FDG PET, but patients did not have to fast before tracer injection. Patients received approximately 200 MBq of 18F-FES. Between FES PET and FDG PET a minimum interval of 24 hours was required.

Baseline Imaging Analysis

Contrast enhanced CT scans were analyzed according to the RECIST 1.1 guidelines (7). All measurable lesions were recorded by the same experienced radiologist (CZ). The FDG PET scans were quantitatively analyzed for all bone lesions by the same nuclear physician, and qualitatively for all non-bone lesions (AG). FES uptake in tumors was

(8)

77

quantified for all lesions seen on CT and FDG PET, as well as for focal non-physiological uptake visible above background with an SUV >1.5 (CV) and randomly checked (AG). In line with previous studies, we used the SUVmax to

calculate tumor FES uptake (9,12,16). We also measured the SUVmean in the

volume of interest (VOI) defined by a 50% isocontour relative to the hottest pixelof the tumor. Due to physiological uptake of the tracer in the liver, liver lesions were excluded from FES PET analyses. The same measurements were performed for the baseline FDG PET. FES SUVmax above 1.5 was

considered FES positive, and FES SUVmax below 1.5 negative in line with

previous studies (9).

The FES/FDG ratio at baseline was established per lesion and divided in 4 groups, based on Kurland et al. (11). Group 1: high FES/high FDG (FES SUVmax>1.5/FDG SUVmax>5.0); group 2: low FES/high FDG, assumed

aggressive behavior (FES SUVmax<1.5/FDG SUVmax>5.0), group 3: high

FES/low FDG, assumed indolent behavior (FES SUVmax>1.5/FDG SUVmax<5.0),

and group 4: low FES/low FDG, but visible on CT scan or bone scan (FES SUVmax <1.5/FDG SUVmax<5.0). Per patient qualitative analyses of the FES

PET and FDG PET scans based on the FES/FDG radio based on Kurland et al. similar to the per lesion analysis. The median FES/FDG ratio was established in three individual lesions per patients, and divided in above mentioned groups.

The heterogeneity score per patient for FES uptake was defined as percentage of all FDG positive lesions that are also positive for FES, with a cut-off value for FES uptake at SUVmax 2.0 (groups: complete concordance,

partial concordance, no concordance)(12).

Response Imaging Analysis

Response assessment on CT was determined anatomically per lesion after eight weeks of treatment according to RECIST 1.1 (8). In short, response was defined as ≥30% reduction in tumor diameter, and progression as a ≥20% increase in diameter. Lesion that showed neither response nor progression were considered to be stable. In case of bone lesions, which cannot be evaluated by CT, response was determined by FDG PET, with progression defined as ≥30% increase in SUV corrected for background, in analogy with the PERCIST 1.0 criteria (17). Response was defined on FDG PET as ≥30% decrease in SUV corrected for background, and lesions that

(9)

78

showed neither response nor progression were considered to be stable. Background correction was applied using a VOI at the unaffected contralateral site whenever available, or at the surrounding tissue of the same origin and subtracted from the SUV of the tumor (i.e. lesion SUVmax/mean minus background SUVmax/mean) resulting in background

corrected SUVmax/mean. Circulating Tumor DNA

Patients could give additional consent for circulating tumor DNA analysis. When consent was obtained, an additional blood sample was taken at baseline. Analysis of all samples was performed at the end of the study to minimize assay variability. EDTA blood samples were centrifuged to isolate circulating tumor DNA and plasma was analyzed for ESR1 mutations by next generation deep sequencing using the commercial Oncomine Breast cell free nucleic acid panel with a detection limit of 0.1% (Thermo Fisher Scientific) (20).

Statistics

The primary endpoint was the relation between tracer uptake on FES-PET to response after eight weeks per lesion. Secondary objectives were 1. to relate mutations found in circulating tumor DNA at baseline to FES uptake in the tumor, 2. to relate FDG uptake at baseline to response 3. to relate the ratio between FDG baseline uptake and FES uptake to response and 4. to relate mutations in circulating tumor DNA to response to treatment. Correlations between baseline uptake, mutations, and response at eight weeks were calculated using a Pearson’s correlation test.

RESULTS Patients

Fifteen women with metastatic breast cancer were included in the trial between October 2016 and March 2017. Median age was 47 years (range 35 - 75). 67% of the patients had both visceral and bone lesions, whereas 33% of the patients had only bone lesions. All but one patient had received prior systemic therapy in the metastatic setting (see Table 1 for more demographic details).

(10)

79

All patients had primary ER-positive breast cancer. Tumor histology of a new biopsy of a metastatic site was available in 14 patients. It was not possible to obtain a new biopsy in one patient, as it was considered not safe to biopsy a thoracic spine metastasis. The new tumor biopsies of the metastases were IHC positive for ER expression in 13 out of 14 patients. The other patient had a previous biopsy of a bone metastasis that was ER positive, but a new biopsy of a liver lesion taken prior to FES PET was negative for ER.

Table 1. patient demographics

Characteristics Number Percentage

Age: median years 47 (range 35-75)

Body mass index (kg/m²) 30.6 (range 21.4 – 46.3) Primary tumor characteristics: ER+/HER2- 15 100% Metastatic tumor characteristics ER+/HER2- ER-/HER2- 13 1 Previous treatment None Hormonal (1-2) Hormonal (>2) Chemotherapy (1-2) Chemotherapy (>2) 0 13 1 2 0 0% 87% 7% 14% 0% Measurable disease Liver Lymph nodes Soft tissue 10 7 3 1 66% 47% 20% 7%

Treatment Result Of Palbociclib And Letrozole

All patients received palbociclib plus letrozole. The median follow-up for patients was 28 weeks (range 8 to 57 weeks). At data cutoff, November 1st 2017, five patients were still on treatment.

(11)

80

In total 40 adverse events were registered in 13 patients, of which 14 adverse events were considered to be related to palbociclib. The most common adverse event was neutropenia and leukopenia (20%). Other adverse events related to palbociclib included epistaxis (7%), fatigue (14%), and different taste sensation (7%). No grade 4 hematological adverse events occurred. One serious adverse event, a hospital admission due to cholecystitis, occurred during the study, but was considered not related to study treatment. The patient with this serious adverse event, did show response on treatment according to RECIST 1.1, however she wished to discontinue treatment. The dose of palbociclib was interrupted for 1 week in 6 patients and reduced to 100 mg according to protocol in three patients due to grade 3 neutropenia.

Figure 2

Examples sagittal fusion PET/CT scans (A,C,E) and maximum intensity projection PET images (B,D,F) of a MBC patient that responded to palbociclib plus letrozole treatment. A and B show the baseline 18F-fluoroestradiol (FES) PET with physiological uptake in the liver, small intestines and the urinary tract. Pathological uptake is seen in nearly all vertebrae, pelvic bones and femora and axillary lymph

nodes. The baseline 18F-fluorodeoxyglucose (FDG) PET is depicted C and D with

physiological uptake in the brain and urinary tract. Pathological uptake was also seen in nearly all vertebrae, pelvic bones, and axillary lymph nodes (right side). After 8 weeks the FDG PET still shows physiological uptake, but pathological uptake is no longer present throughout the vertebrae (E and F). The patient has been on treatment for more than 57 weeks.

Of the 15 patients, 14 were evaluable for the primary endpoint. One patient was not evaluable due to rapid onset- and progression of meningeal metastases. Declining condition of the patient intervened with response

(12)

81

evaluation and subsequent analyses. Ten patients had, in addition to bone metastases, also measurable disease according to RECIST 1.1 (liver metastases n = 6, lymph node metastases n = 2, liver and lymph node metastases n = 2). Four patients had non-measurable, bone only disease. According to RECIST 1.1 and PERCIST 1.0 criteria, five out of 14 patients had progressive disease within eight weeks.

Figure 3

Examples of maximum intensity projection PET images (upper row) and transversal fusion PET/CT scans (lower row) of a patient that did not respond to palbociclib plus letrozole treatment. Image A show the baseline 18F-fluoroestradiol (FES) PET with physiological uptake in the liver, small intestines and the urinary tract. Pathological uptake is only seen in one mediastinal lymph node. The baseline 18

F-fluorodeoxyglucose (FDG) PET is depicted in image B with physiological uptake in the brain, heart, urinary tract, and throughout the colon. Pathological uptake was also seen in several vertebrae in the liver, and pleural lesions. After 8 weeks the FDG PET still shows physiological uptake, pathological uptake is increased in bone lesions, and new liver lesions were visible (image C).

FES PET Analysis In Relation To Response

Based on CT and FDG PET, a total of 279 lesions were identified in 14 patients: 237 bone lesions, and 42 soft tissue lesions. In total, 229 FES positive lesions were observed in 14 patients, and 50 FES negative lesions were found (Table 2 and Supplemental Table 1) in seven patients. Examples of baseline PET images and a follow up FDG PET scans are depicted in Figure

(13)

82

2 and 3. Of the 50 FES negative lesions, 60% showed progression versus 12% of the FES positive lesions (p < 0.01) (Figure 4, Table 2). Of the 50 FES negative lesions, 13 lesions showed metabolic stable disease according to PERCIST, with a change of FDG uptake between -7 to +27%. Only 7/50 (14%) of FES negative lesions did show a metabolic response to treatment at 8 weeks.

Figure 4

Waterfall plot. The Y-axis depicts the percentage of change in FDG PET uptake after 8 weeks of treatment with letrozole and palbociclib, as compared to baseline. Blue bars represent FES positive lesions (SUV > 1.5) and red bars represent FES negative lesions.

Table 2. Cross table of FES PET analysis per lesion and response Lesions Response Stable Progression

FES positive (SUVmax >1.5) 107 95 27 229 FES negative (SUVmax <1.5) 7 13 30 50 114 108 57 279

(14)

83

FES PET Analysis Per Tumor Site

Bone lesions

In total, 237 bone lesions were analyzed, of which 41 (17%) were FES negative. 51% of the FES negative bone lesions showed metabolic progression versus 11% of the FES positive lesions. Only 17% of the FES negative lesions showed metabolic response versus 88% of the FES positive lesions (Supplemental Table 1). The intensity of FES uptake in bone lesions was not correlated to FDG reduction between baseline and 8 weeks (R² = 0.10, p < 0.01).

Visceral lesions

In total 42 visceral lesions were analyzed, of which 9 (21%) were FES negative. All FES negative lesions showed progression (100%), whereas only 12% of the visceral FES positive lesions progressed (Supplemental Table 2). No FES negative lesions showed response, versus 12% of the FES positive lesions. FES uptake in visceral lesions was correlated to response on CT scan (R² = 0.38 P = 0.01).

FDG PET Analysis and FDG/FES uptake ratio per lesion

Quantitative analysis of the baseline FDG uptake per lesion did not show that uptake was related to response at 8 weeks. The ratio between FDG SUVmax at baseline and FES SUVmax was not related to response at 8 weeks

either.

Qualitative analysis per lesion did show a better response rate in lesions with a more indolent behavior (high FES/ low FDG) (61%) versus lesions with an aggressive behavior (low FES/ high FDG) (19%) (Table 3, Supplemental Table 3 and 4).

Per patient analysis

Heterogeneity in lesion uptake of FES was seen between patients and across lesions within individual patients (Figure 5). With the cutoff for SUVmax at 1.5, seven patients had both FES positive and FES negative lesions.

SUVmax varied widely between lesions (median 3.4, range 0.6 – 13.7) and

per patient (median 4.1, range 1.2 – 7.5). Of these seven patients with both FES positive and FES negative lesions, six patients showed a heterogeneous response between lesions.

(15)

84

Table 3. Per lesion analysis according to Kurland et al. Group 1 with median

FES high (SUVmax >1.5) and FDG high (SUVmax >5.0) lesions, group 2: FES low

(SUVmax <1.5)/FDG high lesions with presumed aggressive behavior, group 3:

FES high/FDG low (SUVmax <5) with presumed indolent behavior, and group

4: FES low/FDG low, but visible on conventional imaging. Response at 8 weeks according to PERCIST or RECIST 1.1.

Response Stable Progression

Group 1 (high FES/high FDG) 53 66 21 140

Group 2 (low FES/high FDG) 5 8 13 26

Group 3 (high FES/low FDG) 54 29 6 89

Group 4 (low FES/low FDG) 2 5 17 24

Low median FDG-uptake per patient at baseline did not show a significant correlation with progression free survival (R² = 0.25 p = 0.08). No cut-off value for response prediction with baseline FDG uptake could be found with a receiver operating characteristic curve. The area under the curve was 0.49 for SUVmax, 0.45 for SUVpeak and 0.50 for SUVmean.

Qualitative analysis per patient according to Kurland et al (11). did not result in a significant relation with progression free survival, although PFS in this small study was lower in patients in the group with aggressive behavior (low FES/high FDG) than in the group with more indolent behavior (high FES/low FDG) (Table 4). No patients had median low FES/ low FDG uptake only.

Table 4. Per patient analysis according to Kurland et al. Group 1 with

median FES high (SUVmax> 1.5) and median FDG high (median SUVmax>5.0) in

three lesions, group 2: FES low (median SUVmax<1.5)/FDG high lesions, and

group 3: FES high/FDG low (median SUVmax<5).

Number of patients Median Time to progression (range)

Group 1 (high FES/high FDG) 5 8 weeks (8 – 53)

Group 2 (low FES/high FDG) 5 24 weeks (8 – 39)

(16)

85

With the heterogeneity score (Table 5) based on Gennari et al. (12), a trend towards longer progression free survival was seen with 100% concordance (median progression free survival 48 weeks) versus partial concordance (19 weeks) or no concordance at all (8 weeks).

Table 5. Per patient heterogeneity analysis according to Gennari et al.:

percentage of all FDG positive lesions that are also positive for FES, related to time to progression.

Number of patients Median time to progression (range) Complete concordance (FES+/FDG+=100%) 5 48 weeks (8 – 57) Partial concordance (FES+/FDG+ = 25-50%) 8 19 weeks (8 – 39)

No concordance (FES+/FDG+= 0%) 1 8 weeks

Circulating Tumor DNA Analysis

Eleven patients gave additional consent for circulating tumor DNA biomarker research. Ten patients had circulating tumor DNA, of which six showed mutations in the ESR1 gene. The D538G mutation was found in five samples, the E380Q mutation in four samples, the Y537N in two samples and the S463P mutation in one sample.

Median FES uptake at baseline was not different between patients with an ESR1 mutation (SUVmax 4.19) versus patients without an ESR1 mutation

(SUVmax 4.08) (Figure 4). No difference in duration of treatment was found

in patients with an ESR1 mutation (28 weeks) versus patients without an ESR1 mutation (31 weeks).

(17)

86

Figure 5

Distribution of maximum standardized uptake value (SUVmax) per lesion per patient measured by FES PET. Lesions are divided into stable (blue circles), response (green diamonds) or progression (red squares) after 8 weeks. The dashed line indicates the threshold on FES PET for ER positive lesions (above) and ER negative lesions (below). Stars below indicate those patients with samples analyzed for cell free DNA. Blue stars indicate an ESR1 mutation.

DISCUSSION

This is the first feasibility trial to investigate the use of whole body

FES PET as a potential predictive imaging marker for response to

palbociclib and letrozole. This feasibility trial indicates that low FES

uptake on FES PET in MBC lesions is related to a lack of response and

early progression on letrozole and palbociclib treatment, particularly

in visceral lesions.

Good biomarkers for response to treatment with CDK inhibitors are

urgently needed, as palbociclib combined with letrozole or

fulvestrant is available at high costs, and with significant toxicity. In

this setting, prediction of non-response is particularly relevant. To

date the only biomarker for response to palbociclib and letrozole has

(18)

87

been the ER expression. Other biomarkers, including ESR mutations,

have been analyzed but have not shown added value to the ER (4).

Molecular imaging has not been assessed in this setting so far. In our

study, as hypothesized, the majority (60%) of FES negative lesions

showed progression after eight weeks versus 12% of the FES positive

lesions. Not unexpectedly, also FES positive lesions had a high

positive predictive value of 94 percent. The FES PET also allows

analysis of individual lesions and gives insight in ER expression

heterogeneity within a patient. A classification based on this

heterogeneity was proposed by Gennari (12). In our small study, the

measure of ER expression heterogeneity indeed does appear to

affect responsiveness to ER targeting treatment. Together with the

present data, this could support further studies incorporating FES

PET, and the heterogeneity classification, in future larger studies.

With conventional imaging, response measurement in bone

metastases -the most common metastatic site in ER-positive breast

cancer- is notoriously difficult. Bone lesions are only evaluable with

CT scan when a soft tissue component is present, according to the

RECIST 1.1. We have therefore chosen to assess response in bone

lesions according to the PERCIST criteria. The PERCIST criteria are

based on the change in uptake of all types of malignancies;

furthermore they are not specified for bone lesions. In addition, the

PERCIST criteria use a broader bandwidth to call something a true

difference indicating metabolic change (minus or plus 30% of FDG

uptake compared to baseline) than the EORTC criteria (minus 15 to

plus 25%) (21). Using the 1999 EORTC criteria, in our study 2 FES

negative lesions would have shown metabolic progression, rather

than stable according to PERCIST.

This feasibility trial was not powered to analyze a response per

patient. Progressive disease according to RECIST 1.1 was seen in 3

patients with high FES uptake in multiple lesions; however

(19)

88

progression occurred in liver metastases. These lesions were not

evaluable for FES PET analysis. Although per lesion analysis

(performed in the present study) provides the most detailed insight

in response to the treatment, whole body scoring systems are

desirable in order to ultimately predict response per patient. In this

setting, we analyzed our patients according to quantitative analysis

proposals by Kurland et al. to identify tumor lesions that were more

indolent (low FDG uptake and high ER uptake) versus those that were

more aggressive (high FDG uptake and low ER uptake). We found a

trend towards longer progression free survival in those patients with

a more indolent behavior but not significant. This is most likely

related to our limited numbers of patients.

Based on the data by Genarri et al., it might be possible to identify

patients who do not need the combination treatment (likely: patients

with indolent, 100% concordant disease), those who will not respond

to treatment (likely: patients with aggressive, less concordant

disease). Based on the present data, it would make sense to pursue

the role of FES PET and whole body scoring methods further, in larger

trials. Not only FES PET has potential to serve as a biomarker, but

possibly also FDG PET could be evaluated as such. Changes in FDG

uptake after only 2 weeks treatment could predict 8 weeks CT

response with other targeted agents, such as trastuzumab-emtansine

(22). In light of this, it could be of interest to investigate in future

trials whether an early FDG PET scan can predict responses to

palbociclib and letrozole at eight weeks.

We also explored whether mutations in the ESR1 gene could result in

altered FES uptake. We found a significant amount of circulating

tumor DNA in 10 out of 11 patients of which 6 showed mutations in

the ESR1 gene. We were unable to find a correlation between ESR1

mutations and FES uptake or duration of response. It is likely that not

all metastases have an ESR1 mutation, and as the circulating tumor

(20)

89

DNA was extracted from the blood it is not possible to identify which

lesion carries a mutation. However, it is conceivable that in larger

studies, assessment of tumor heterogeneity by means of circulating

tumor DNA can be related to response (23,24).

In conclusion, this feasibility trial indicates that FES PET has a high

negative predictive value in visceral lesions as well as a high positive

predictive value for response of an MBC lesion to letrozole plus

palbociclib. ER expression heterogeneity on FES-PET and ratio of FES

to FDG uptake may provide valuable additional information in this

setting. FES-PET may therefore be a potential biomarker for whole

body response for this combination treatment. This will have to be

confirmed in future clinical trials.

ACKNOWLEDGMENTS

Financial support: Pfizer Oncology the Netherlands provided the

palbociclib and a research grant to the UMCG for translational

procedures (imaging, biopsy, blood assessment).

(21)

90

REFERENCES

1. Cardoso F, Harbeck N, Fallowfield L, Kyriakides S, Senkus E and ESMO Guidelines Working Group: Locally recurrent or metastatic breast cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 2012;23:S7:vii11-19

2. Allred DC, Harvey JM, Berardo M, Clark GM. Prognostic and predictive factors in breast cancer by immunohistochemical analysis. Mod Pathol 1998;11:155-168.

3. Finn RS, Crown JP, Lang I et al. the Cyclin-dependent kinase 4/6 inhibitor palbociclib in combinatiopn 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. Lancet Oncol 2015;16:25-35.

4. Cristofanilli M, Turner NC, Bondarenko I et al. Fulvestrant plus palbociclib versus fulvestrant plus placebo for treatment of hormone receptor positive, HER2 negative metastatic breast cancer that progressed on previous endocrine therapy (PALOMA-3) : final analysis of the multicentre, double blind, phase 3 randomised controlled trial. Lancet oncol 2016;17:425-439.

5. Finn RS, Crown J, Lang I et al. Overall survival results from the randomized phase II study of palbociclib in combination with letrozole versus letrozole alone for frontline treatment of ER+/HER- advanced breast cancer (PALOMA-1; TRIO-18). JClin ONcol 35;2017,suppl abstract 1001

6. www.nice.org.uk/guidance/indevelopment/gid-ta10095; accessed May 2017

7. Simmons C, Miller N, Geddie W, et al. Does confirmatory tumor biopsy alter the management of breast cancer patients with distant metastases? Ann Oncol 2009;20:1499-1504.

8. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, Dancey J, Arbuck S, Gwyther S, Mooney M, Rubinstein L, Shankar L, Dodd L, Kaplan R, Lacombe D, Verwij J. New response evaluation in solid tumours: Revised RECIST guideline (version 1.1) Eur J Cancer 2009;45:228-247.

9. van Kruchten M, de Vries EG, Brown M et al. PET imaging of oestrogen receptors in patients with breast cancer. Lancet Oncol 2013;14:e465-475.

10. Groheux D, Majdoub M, Tixier F, et al. Do clinical, histological or immunohistochemical primary tumour characteristics translate into different [18F]-FDG PET/CT volumetric and heterogeneity features in

(22)

91

stage II/III breast cancer? Eur J Nucl Med Mol Imaging 2015; 42:1682-1691.

11. Kurland BF, Peterson LM, Lee JH et al. Estrogen receptor binding (18 F-FES PET) and glycolytic activity (18F-FDG PET) predict progression-free survival on endocrine therapy in patients with ER+ breast cancer. Clin Cancer Res 2016;407-417.

12. Gennari A, Brain E, Nanni O et al. Molecular imaging with 18 F-Fluoroestradiol (18F-FES) to assess intra-patient heterogeneity in metastatic breast cancer (MBC): a European TRANSCAN program. Ann Oncol 2017;28:suppl 5

13. Angus L, Beije N, Jager A, Martens JWM, Sleijfer S. ESR1 mutations: moving towards guiding treatment decision-making in metastatic breast cancer patients. Cancer Treat Rev 2017;52:33-40.

14. Takeshita T, Yamamoto Y, Yamamoto-Ibusuki M et al. Droplet digital polymerase chain reaction assay for screening of ESR1 mutations in 325 breast cancer specimens. Trans res 2015;166:540-553.

15. Oken M, Creech R, Tormey D et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. AM J Clin Oncol 1982;5:649-655. 16. Linden HM, Kurland BF, Peterson LM et al. Fluoroestradiol positron emission tomography reveals differences in pharmacodynamics of aromatase inhibitors, tamoxifen and fulvestrant in patients with metastatic breast cancer. Clin Cancer Res 2011;17:4799-4805.

17. Wahl RL, Jacene H, Kasamon Y, Lodge MA. From RECIST to PERCIST: evolving considerations for PET response criteria in solid tumors. J Nucl Med 2009;50:S1

18. Hammond MEH, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptor in breast cancer. J Clin Oncol 2010;28:2784-2795.

19. Boellaard R, Delgado-Bolton R, Oyen WJ et al. FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0. Eur J Nucl Met Mol Imaging 2015; 42: 328-354.

20. www.thermofisher.com/nl/en/home/life-science/cancer-

research/cancer-genomics/liquid-biopsy-cancer-research- applications/oncomine-cell-free-dna-assays-liquid-biopsy-clinical-research.html. Accessed Nov 2017

21. Young H, Baum R, Cremerius U et al. Measurement of clinical and subclinical tumour response using 18F-fluorodeoxyglycose and positron emission tomography: review and 1999 EORTC recommendations. Eur J Cancer 1999;35:1773-1782.

(23)

92

22. Gebhart G, Lamberts LE, Wimana Z et al. Molecular imaging as a tool to investigate heterogeneity of advanced HER2-positive breast cancer and to predict patient outcome under transtuzumab emtansine (T-DM1): the ZEPHIR trial. Ann Oncol 2016;27:619-624.

23. Yates LR. Intratumoral heterogeneity and subclonal diversification of early breast cancer. Breast 2017;34:S36-S42

24. Fribbens C, Garcia Murillas I, Beany M et al. Tracking evolution of aromatase inhibitor resistance with circulating tumor DNA analysis in metastatic breast cancer. Ann Oncol 2017; doi: 10.1093/annonc/mdx483

(24)

93

Supplemental Tables

Supplemental Table 1. Cross table of FES PET analysis per bone lesion and

response measured based on PERCIST 1.0 criteria by FDG PET scan at eight weeks.

Bone lesions Metabolic response (≥30% decrease in FDG uptake) Metabolic stable disease (<30% decrease and <30% increase in FDG uptake) Metabolic progression (≥30% increase in FDG uptake) FES positive (SUVmax >1.5) 103 70 23 196 FES negative (SUVmax <1.5) 7 13 21 41 110 83 44 237

Supplemental Table 2. Cross table of FES PET analysis per visceral lesion

and response measured based on RECIST 1.1 criteria by CT scan at eight weeks. Visceral lesions Partial response (> 20% reduction in diameter on CT scan) Stable disease (<20% reduction and <30% increase in diameter on CT scan) Progressive disease (>30% reduction in diameter on CT scan) FES positive (SUVmax >1.5) 4 25 4 33 FES negative (SUVmax <1.5) 0 0 9 9 4 25 13 42

(25)

94

Supplemental Table 3. Per bone lesion only analysis according to Kurland et

al. Group 1 with median FES high (SUVmax >1.5) and FDG high (median

SUVmax >5.0) lesions, group 2: FES low (median SUVmax <1.5)/FDG high

lesions with presumed aggressive behavior, group 3: FES high/FDG low (median SUVmax <5) with presumed indolent behavior, and group 4: FES

low/FDG low, but visible on conventional imaging. Response rate at 8 weeks according to PERCIST criteria.

Bone lesions Metabolic response (≥30% decrease in FDG uptake) Metabolic stable disease (<30% decrease and <30% increase in FDG uptake) Metabolic progression (≥30% increase in FDG uptake) Group 1 (high FES/high FDG) 51 53 19 123 Group 2 (low FES/high FDG) 5 8 7 20 Group 3 (high FES/low FDG) 52 71 4 73 Group 4 (low FES/low FDG 2 5 14 21

(26)

95

Supplemental Table 4. Per visceral lesion only analysis according to Kurland

et al. Group 1 with median FES high (SUVmax >1.5) and FDG high (median

SUVmax >5.0) lesions, group 2: FES low (median SUVmax <1.5)/FDG high

lesions with presumed aggressive behavior, group 3: FES high/FDG low (median SUVmax <5) with presumed indolent behavior, and group 4: FES

low/FDG low, but visible on conventional imaging. Response rate at 8 weeks according to RECIST 1.1 criteria.

Visceral lesions Partial response (> 20% reduction in tumor diameter on CT scan) Stable disease (<20% reduction and <30% increase in diameter on CT scan) Progressive disease (>30% reduction in tumor diameter on CT scan) Group 1 (high FES/high FDG) 2 13 2 17 Group 2 (low FES/high FDG) 0 0 6 6 Group 3 (high FES/low FDG) 2 12 2 16 Group 4 (low FES/low FDG 0 0 3 3

(27)