Assessment of Folate Receptor-α and Epidermal Growth Factor Receptor Expression in Pemetrexed-Treated Non–Small-Cell Lung Cancer Patients

Accepted Manuscript Assessment of Folate Receptor alpha and Epidermal Growth Factor Receptor Expression in Pemetrexed-Treated Non-Small Cell Lung Cancer Patients Daniel C. Christoph , Bernadette Reyna-Asuncion , Biftu Hassan , Cindy Tran , Julia D. Maltzman , Daniel J. O’Shannessy , Thomas C. Gauler , Jeremias Wohlschlaeger , Martin Schuler , Wilfried E. Eberhardt , Fred R. Hirsch PII:

S1525-7304(14)00109-0

DOI:

10.1016/j.cllc.2014.05.002

Reference:

CLLC 278

To appear in:

Clinical Lung Cancer

Received Date: 11 February 2014 Revised Date:

8 April 2014

Accepted Date: 19 May 2014

Please cite this article as: Christoph DC, Reyna-Asuncion B, Hassan B, Tran C, Maltzman JD, O’Shannessy DJ, Gauler TC, Wohlschlaeger J, Schuler M, Eberhardt WE, Hirsch FR, Assessment of Folate Receptor alpha and Epidermal Growth Factor Receptor Expression in Pemetrexed-Treated NonSmall Cell Lung Cancer Patients, Clinical Lung Cancer (2014), doi: 10.1016/j.cllc.2014.05.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Assessment of Folate Receptor alpha and Epidermal Growth Factor Receptor Expression in Pemetrexed-Treated Non-Small Cell Lung Cancer Patients Daniel C Christoph1, Bernadette Reyna-Asuncion1, Biftu Hassan1, Cindy Tran1, Julia D Maltzman2,

Eberhardt3, and Fred R Hirsch1 1

RI PT

Daniel J O’Shannessy2, Thomas C Gauler3, Jeremias Wohlschlaeger4, Martin Schuler3, Wilfried E

Department of Medicine, Division of Medical Oncology, University of Colorado Denver, Aurora, CO,

USA; 2Morphotek Inc., Exton, PA; 3Department of Medical Oncology, West German Cancer Center,

SC

University Hospital Essen, University Duisburg-Essen, Essen, Germany; 4Department of Pathology and Neuropathology, West German Cancer Center, University Hospital Essen, University Duisburg-Essen,

M AN U

Essen, Germany

Source of Funding:

This work was supported by a research grant from Morphotek, Inc., a subsidiary of Eisai Corporation

TE D

of North America.

Running title:

Corresponding author:

EP

FRA & EGFR as molecular targets

AC C

Fred R Hirsch, MD, PhD; Department of Medicine, Division of Medical Oncology, University of Colorado Denver, 12801 E. 17th Avenue, RC-1 South, Mail Stop 8117, Aurora, CO 80045; Tel: (303)724-3858, Fax: (303)724-3162; e-mail: [email protected]

Word count abstract:

226

Word count manuscript:

4,000

Word count clinical practice points:

248 1

ACCEPTED MANUSCRIPT Conflicts of interest: D.C.C. received travel support from Lilly Germany. J.D.M. and D.J.O. are employees of Morphotek, Inc., a subsidiary of Eisai, Inc.. T.C.G. received consulting fee or honorarium and travel support from Lilly

RI PT

Germany. M.S. has received travel support from Lilly Germany. W.E.E. received research funding, consulting fee and honoraria from Lilly Germany. F.R.H. received consulting fee from Eli Lilly and grant support from Morphotek Inc., a subsidiary of Eisai, Inc. (through University of Colorado). All

SC

other authors state that they have no conflicts of interest to declare.

M AN U

MicroAbstract:

Agents targeting the Folate Receptor Alpha (FRA) or Epidermal Growth Factor Receptor (EGFR) are approved or in clinical development. FRA and EGFR expression in advanced Non-Small Cell Lung Cancer (NSCLC) was evaluated. In 160 advanced NSCLC patients, 29% had tumors expressing high

TE D

levels of both the receptors. This subgroup would be candidates for clinical trials studying combined

AC C

EP

targeted therapies.

2

ACCEPTED MANUSCRIPT

Abstract Introduction: Folate receptor alpha (FRA) regulates cellular uptake of folates and antifolates (e.g. pemetrexed) and is frequently expressed in pulmonary adenocarcinoma. Epidermal Growth Factor

RI PT

Receptor (EGFR) is an established therapeutic target in Non-Small Cell Lung Cancer (NSCLC). Therapies targeting FRA or EGFR are available. The association between FRA and EGFR expression in advanced NSCLC has not been explored. Combining therapeutic FRA-antibodies with an EGFR

SC

inhibitor might be beneficial, if both of the targets are significantly co-expressed. Patients and Methods: Specimens from 160 advanced NSCLC patients receiving pemetrexed-based chemotherapy

M AN U

were assessed for membranous FRA and EGFR protein expression by immunohistochemistry using the H-score. EGFR (exons 18-21) and K-Ras (exon 2) mutations were determined. Results were correlated to patients’ clinicopathological data, progression-free survival (PFS) and overall survival (OS). Results: 47 patients (29%) had tumors with strong FRA and EGFR expression, but no statistically

TE D

significant correlation was seen between protein levels of FRA and EGFR. High membranous FRA expression (H-score ≥20) was associated with prolonged PFS (5.5 vs. 3.4 months, HR=0.6060, P=0.0254) and improved OS (12.1 vs. 6.4 months, HR=0.5726, P=0.0076). Conclusion: Survival

EP

times are improved in NSCLC patients whose tumors show strong membranous FRA expression. No statistical correlation between membranous FRA and EGFR expression was demonstrated in advanced

AC C

NSCLC, but 29% of patients had higher expression of both the receptors and could be suitable for combined targeted therapies.

Key words:

correlation of receptor expression, antifolate chemotherapy, predictive biomarker, therapeutic antibody, combination of targeted agents

3

ACCEPTED MANUSCRIPT

Introduction Lung cancer continues to be the leading cause of cancer deaths worldwide. Non-Small Cell Lung Cancer (NSCLC) accounts for approximately 85% of all lung cancer cases1. A platinum-based

RI PT

combination chemotherapy with or without bevacizumab is currently the standard first-line therapy for patients with advanced Epidermal Growth Factor Receptor (EGFR) and Anaplastic Lymphoma Kinase (ALK) wild-type NSCLC2. The multitarget antifolate pemetrexed is used in combination with cisplatin3

SC

or carboplatin4 in non-squamous NSCLC, and is often also administered after platinum-based chemotherapy as continuous maintenance therapy, as single agent after progression of first-line

M AN U

therapy5,6 or given as first-line therapy to patients who are medically unfit for platinum-based combination chemotherapy7.

Three transporters are identified for the transport of folates and antifolates into eukaryotic cells: the ubiquitous reduced folate carrier (RFC), the proton-coupled folate transporter (PCFT), and the

TE D

folate receptors of which Folate Receptor Alpha (FRA) being the most widely studied isoform8,9. Unless the receptor is highly expressed relative to RFC, and transport mediated by RFC and other routes are impaired, FRA-mediated transport contributes little to the transfer of either folates or

EP

antifolates through the cell membrane10. FRA is a glycosylphosphatidylinositol (GPI) anchored cell surface protein that binds free folate with high affinity11 and assists in the uptake of antifolates via a

AC C

classical mechanism of receptor-mediated endocytosis. Often, detectable amounts of FRA protein are expressed in lung tumors that retain alveolar epithelial cell characteristics, such as adenocarcinoma and large cell carcinoma12,13.

Recently novel therapeutic agents directed against FRA were developed and they are currently under clinical investigation for the NSCLC treatment. These agents are distinguished into two different classes based on their mechanism of action: they are either conjugates of a cytotoxic agent with folic acid (e.g. vintafolide, Endocyte, Inc., West Lafayette, IN; Merck Sharp & Dohme Corp. is continuing

4

ACCEPTED MANUSCRIPT the clinical development)14 or antibodies eliciting robust antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity (e.g. farletuzumab, Morphotek, Inc., Exton, PA)15,16. EGFR is a promising therapeutic target in NSCLC17. The EGFR-directed Tyrosine Kinase

RI PT

Inhibitor (TKI) erlotinib is also an established treatment option for patients with advanced disease who have been pre-treated with platinum-based combinations18,19. Patients with EGFR-mutant NSCLC have a high likelihood of clinical benefit from EGFR-TKI. Also 1-7% of patients with unknown or wild-type

SC

EGFR status respond to EGFR TKI in first-line therapy and 3-7% in second/third-line therapy20. EGFR expression evaluated by immunohistochemistry (IHC) using an antibody that detects the intracellular

M AN U

domain (ID) of the EGFR predicted response and clinical outcome (progression-free survival (PFS), overall survival (OS)) in NSCLC patients receiving an EGFR TKI21. Of note, an ID-specific EGFR antibody also detects truncated forms of the EGFR that are constitutively active21. Furthermore, cetuximab, an anti-EGFR monoclonal antibody, was studied in combination with cisplatin and vinorelbine for the treatment of metastatic NSCLC22. In a post-hoc subset analysis EGFR protein

was predictive of survival22,23.

TE D

expression, detected by IHC using an antibody binding to the extracellular domain (ED) of the EGFR,

Although the combination of targeted drugs (e.g. bevacizumab with erlotinib) is feasible24,

EP

efficacy so far has been disappointing in unselected patient populations25. However, the interest in

AC C

additional, beneficiary combinations of targeted drugs is increasing. E.g. onartuzumab, a monoclonal antibody against the Met receptor, is currently being investigated in combination with erlotinib in patients with Met receptor positive NSCLC26. Before testing and establishing novel combinations of targeted agents, both of the targets need to be present in the tumors. Due to the mechanism of action and the safety profile, immunotoxins or therapeutic antibodies directed against FRA might act synergistically in combination with EGFR TKIs, most likely if FRA and EGFR are expressed at higher levels, but this is a putative assumption The anti-tumor activity of farletuzumab combined with EGFRTKIs has not yet been tested in vitro. For in vitro tests using farletuzumab, addition of 20% human 5

ACCEPTED MANUSCRIPT serum is necessary15, and such a high concentration of serum excludes any meaningful investigation of TKIs in this setting. Furthermore, it has been reported that FRA and EGFR are enriched in specialized plasma membrane microdomains known as caveolae, which are membrane invaginations at the cell

RI PT

surface and characterized by high concentrations of gangliosides, sphingomyelin, and cholesterol, and the protein caveolin27,28.

Because therapeutics targeting FRA or EGFR are available, we wanted to study the prevalence

SC

of high FRA and EGFR expression in advanced NSCLC. A potential correlation between expression of the two receptors could serve as a basis for combinational therapeutic strategies. We have recently

M AN U

reported that FRA is frequently expressed in advanced pulmonary adenocarcinomas29 of patients undergoing pemetrexed-based chemotherapy. Furthermore, we and other groups observed that FRA is expressed in the membrane and cytoplasm of large cell carcinomas and squamous cell carcinomas, but at significantly lower levels compared to adenocarcinomas13,29,30. Therefore, we evaluated the expression of FRA and EGFR in tumors from pemetrexed-treated NSCLC patients and correlated the

TE D

expression levels of FRA and EGFR to clinicopathological data.

Patients and Methods

EP

Patient cohort

The study included patients of the West German Cancer Center between November 2004 and October

AC C

2011, who received pemetrexed-based chemotherapy as first or further line treatment for metastatic NSCLC and whose tissue specimens were available for EGFR protein detection. If possible, EGFR/KRas mutational diagnostics using remaining specimens was performed. For this study sufficient pretreatment tumor samples for EGFR IHC was available from 160 patients. Clinicopathological data such as age, gender, histology, complete history and radiographic findings were collected (Table 1). As proposed by the International Association for the Study of Lung Cancer (IASLC) tumor staging was based on the tumor, node, and metastasis (TNM) staging system 6

ACCEPTED MANUSCRIPT (7th edition)31. TNM and IASLC stages refer to the initial diagnosis of NSCLC. For the evaluation of response of NSCLC patients to treatment Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 was used32. PFS was defined from the first day of chemotherapy treatment until progression

RI PT

or the last visit where a patient was alive without progression. OS was calculated as the time between the start of pemetrexed-based chemotherapy until the date of death, or the date of last follow-up. At the last follow-up patients were censored if they were still alive or lost to follow-up. On December 31,

SC

2013 surveillance of PFS and OS for this study was stopped. For the tissue used in this study, living patients provided written informed consent. The Ethics Committee of the Medical Faculty of the

DNA extraction and mutational analysis

M AN U

University Duisburg-Essen approved this retrospective study (no. 10-4404).

Pre-treatment archival tissue specimens were used for genomic analysis. Genomic DNA was isolated from the formalin-fixed, paraffin-embedded (FFPE) tissue blocks. After macrodissection, the QIAamp DNA FFPE Tissue Kit (Qiagen) was used for DNA extraction according to the manufacturer’s

TE D

instructions33. Polymerase chain reaction (PCR) was performed to amplify the exons 18 to 21 of the EGFR gene or exon 2 of the Kirsten RNA-associated rat Sarcoma 2 virus (K-Ras) gene with the following

primers

(forward

and

reverse,

respectively)34,35:

EGFR

exon

18

(5′-

EP

TTGTCCTTCCAAATGAGCTG-3′ and 5′-GAAAAACACTGGAGTTTCCC-3′, amplicon size of 191

AC C

bp), EGFR exon 19 (5′-ATAATCAGTGTGATTCGTGG-3′ and 5′-TTATACCCACTAGAGCTAGA3′; amplicon size of 249 bp), EGFR exon 20 (5′-CACTGCATCTGTCACTTCAC-3′ and 5′GCAAACTCTTGCTATCCCAG-3′,

amplicon

size

of

234

bp),

EGFR

exon

21

(5′-

GAGAAAAGTTAATGGTCAGC-3′ and 5′-CTCACCCAGAATGTCTGGAG-3′; amplicon size of 255 bp),

and

K-Ras

exon

2

(5′-GGTGAGTTTGTATTAAAAGG-3′

and

5′-

CAGATAACTTAACTTTCAGC-3′; amplicon size 122). Amplifications and direct sequencing were done by CD Genomics (Shirley, NY, USA). In some cases, EGFR and K-Ras mutational diagnostics

7

ACCEPTED MANUSCRIPT were instead performed at the Department of Pathology and Neuropathology of the University Hospital Essen (Essen, Germany).

RI PT

Tissue microarray (TMA) preparation and IHC for FRA and EGFR The preparation of TMA was described previously29. The TMA used for this study was based on pretreatment archival tissue and contained 3 cores of each FFPE tumor tissue block with a size of 1 mm.

SC

Fresh cuts of the TMA blocks at 4 µm thickness were mounted on PlusGOLD® slides (Thermo Scientific Fisher, Inc.) for IHC staining. In a few cases the cores were lost from TMA sections or tumor

M AN U

samples were inadequate for TMA preparation. In these cases the whole tissue sections of the original FFPE tumor blocks were used.

The IHC detection of FRA protein has been described elsewhere29. Briefly, deparaffinization, rehydration, heat-induced antigen-retrieval, and IHC staining process was performed in the Dako Autostainer Universal Staining System (Dako). After blocking for endogenous peroxidase activity,

TE D

unconjugated antibody, mouse monoclonal FRA (clone 26B311, dilution 2.5 µg/ml, Morphotek, Inc.), was applied to the tissue sections and incubated for 60 minutes at room temperature (RT). For negative control slide, non-immunized unconjugated universal negative control mouse antibody (Dako), was

EP

used and incubated under the same conditions. MACH4 Mouse Probe Primary Antibody Enhancer

AC C

(Biocare Medical) was added and incubation for 15 minutes was followed by addition of Polymer-HRP reagent (Biocare) for 20 minutes at RT. After development of visual color with 3-3’-diaminobenzidine (DAB) tetra-hydrochloride (Dako), slides were counterstained with Hematoxylin (Dako). Human kidney sections were used as positive control for FRA IHC because FRA is expressed in the proximal tubules of the kidney13,30. Figure 1A displays highly specific staining of the FRA antibody used, since FRA expression is restricted to the luminal surface of proximal tubule cells36,37. We have previously published the IHC procedure for EGFR21. Briefly, slides underwent deparaffinization in a dry oven at 60°C for an hour. Antigens were retrieved by treating with Cell 8

ACCEPTED MANUSCRIPT Conditioning 1 (Ventana Medical Systems Inc.) for 60 minutes in a Benchmark XT automated stainer (Ventana). Slides were incubated with pre-diluted EGFR antibody (clone 5B7, specific antibody concentration of ≈0.4 µg/ml, Ventana) at 37° C for 16 minutes. For negative control, unconjugated

RI PT

non-immunized ready-to-use mouse antibody (Ventana) was applied as the primary antibody. Staining visualization was accomplished with ultraView Universal DAB detection kit (Ventana) with an extra washing step. After counterstaining with Hematoxylin (Ventana) for 4 minutes and post-

SC

counterstaining with Bluing (Ventana) for 4 minutes, slides were washed with mild soapy water and then dehydrated in ethyl alcohol (Leica Microsystems) and xylene (Leica) baths before applying

M AN U

coverslips. Tissue sections of a known EGFR expressing NSCLC sample were included as positive controls.

IHC assessment was performed independently by a pathologist (B.R.-A.) and a trained reader (D.C.C.), and ambiguous cases were reassessed jointly until a consensus was reached. Furthermore, a round-robin test for FRA IHC on a subset of patients was performed with the Laboratory Corporation

TE D

of America (Los Angeles, CA, USA). Images were captured at 400-fold magnification and immunoreactivity levels in each section were evaluated under a light microscope. Evaluation of FRA or EGFR protein expression was based on membranous staining13,21,29. Tumor cell staining intensity was

EP

graded on a scale from 0+ to 3+. The H-score was derived by adding the products of the percentage of

AC C

positive cells (0-100%) and the intensity (0-3) for that group, resulting in a final H-score ranging from 0 to 30013,29. Three different cores of tumor samples were taken from each patient, and the mean Hscore was used in the final analysis. Tumor samples representing different H-scores for either FRA or EGFR are displayed in Figure 1. Statistics The objective of this study was to assess a correlation between FRA and EGFR expression in advanced NSCLC. Another goal was to study associations between FRA and EGFR protein levels with clinicopathological data and outcome for NSCLC patients treated with pemetrexed, which is 9

ACCEPTED MANUSCRIPT transported within tumor cells by the FRA. Descriptive analyses were performed on the H-scores and correlated to patient's clinical data: age at diagnosis, gender, histological grade or subtype (adenocarcinoma vs. large cell carcinoma vs. squamous cell carcinoma), and TNM/IASLC stage. A

RI PT

Spearman's test38 was used to further correlate as continuous variables membranous FRA and EGFR Hscores, and age at diagnosis. Distributional differences of H-scores between different levels of categorical variables (gender, histological subtype or grade, stage, EGFR or K-Ras mutational status, response, or clinical disease stabilization) were analyzed using either the Kruskal-Wallis test or

SC

Wilcoxon ranked sum test (which is equivalent to Kruskal-Wallis test when comparing two groups)39.

M AN U

For the Kaplan-Meier (KM) curves of PFS and OS, we tested binary H-score cutoff-values of biomarkers and applied the maximum chi-square method30,40,41. Univariate and multivariate analyses of both PFS and OS were performed using the Cox proportional hazards model. A P-value less than 0.05 was considered statistically significant. Statistical analyses were performed by using GraphPad Prism

TE D

(Version 5.00 for Windows, GraphPad Software) or SPSS software package (Version 21, SPSS Inc.).

Results Patient cohort

EP

Demographic and clinicopathologic data of the patients are summarized in Table 1. Median PFS and OS were 4.9 months (95%-CI: 4.1-5.7) and 10.0 months (95%-CI: 7.7-12.4), respectively. The main

AC C

histologic subtype in the present cohort was adenocarcinoma, occurring in 123 patients (77%); large cell carcinoma and squamous cell carcinoma subtypes were found in 9 (6%) and 8 (5%) patients, respectively. Gender was equally distributed, 74 (46%) patients were females and 86 (54%) were males. Pemetrexed-based chemotherapy was administered to 54 (34%) patients as first-line treatment. During their course of disease 90 patients (56%) of the entire cohort got an EGFR TKI as single-drug. .

10

ACCEPTED MANUSCRIPT Intratumoral correlation of FRA and EGFR expression Before we assessed the membranous protein levels of FRA, we performed a round-robin test with the Laboratory Corporation of America. The FRA H-scores determined by our laboratory and the H-scores

RI PT

evaluated by the reference laboratory were strongly correlated: Spearman’s correlation was r=0.9413 (P=0.0002).

The median H-score for membranous FRA expression was 62.5 (range: 0-275). Assessment of

SC

membranous EGFR expression revealed a median H-score of 100 (range: 0-240). Data about protein expression of the receptors are displayed in Figure 2A and summarized in Suppl. Table 1. Furthermore,

M AN U

47 patients (29%) had tumors with strong membranous expression of both FRA and EGFR receptors (Suppl. Table 2) meaning expression higher than the median of each receptor. However, there was no correlation between membranous FRA and EGFR protein expression (Figure 2B, Spearman’s correlation; r=0.1136; P=0.1527).

Association between clinicopathological data, response, outcome and FRA expression

TE D

A trend for a higher membranous FRA expression in females (mean 93±9, median 80) compared to males (mean 71±7, median 45) was observed, but it did not reach statistical significance (p=0.0769). Membranous expression of FRA was significantly higher in adenocarcinomas (mean 86±7, median 80)

EP

compared to large cell carcinomas (mean 36±13, median 35) or squamous cell carcinomas (mean

AC C

44±29, median 0) (PKW=0.0430; Figure 3A). No significant association between membranous FRA protein levels and histological grade, T-, N-, M- or IASLC-stages, EGFR- or K-Ras mutational status were observed (Table 2A). No association between response (complete and partial remission, P=0.6286) or clinical disease control (complete or partial remission and disease stabilization, P=0.4613) and membranous protein levels of FRA was found (Suppl. Table 3). High membranous FRA expression (H-score ≥20, maximum χ2 of 4.993) was associated with a prolonged PFS (5.5 vs. 3.4 months, P=0.0254, HR=0.6060; 95%-CI: 0.3906-0.9403; Figure 4A). Similar, high expressors (H-score ≥20, maximum χ2=7.133) showed an improved OS (12.1 vs. 6.4 11

ACCEPTED MANUSCRIPT months, P=0.0076, HR=0.5726, 95%-CI: 0.3803-0.8621; Figure 4B). We performed survival analyses for the subgroups of patients who received first-line treatment only (Figures 4C/D) or second-line therapy only (Figures 4E/F). These analyses revealed similarly significant results.

RI PT

At univariate analysis of the Cox proportional hazards model, OS was significantly associated with membranous protein levels of FRA, the histological subtype or K-Ras mutational status (PCOX=0.045, PCOX=0.029, or PCOX=0.049, respectively). Furthermore, a trend for a significant

SC

association between OS and EGFR mutational status or the administration of EGFR TKI was observed (PCOX=0.053 or PCOX=0.078, respectively). All of the five parameters were included in the multivariate

M AN U

analysis resulting in a borderline significant association between membranous FRA expression and OS (PCOX=0.079).

Association between clinicopathological data, response, outcome and EGFR expression No significant association between membranous EGFR protein expression and histological subtype or grade, T-, N-, or IASLC-stages, or K-Ras mutational status were noted (Table 2B). Expression of

TE D

EGFR in tumors presenting with distant metastases at diagnosis (mean 101±6, median 100) was higher compared to locally advanced stages (mean 82±9, median 80), but this difference was not significant (P=0.0820). Similarly, EGFR mutated tumors expressed more EGFR protein (mean 126±15, median

AC C

(P=0.0750).

EP

120) compared to wild-type tumors (mean 94±6, median 100), but the difference was not significant

Membranous expression of EGFR was not associated with PFS (cut-off H-score of ≥110, maximum χ2 of 2.643, P=0.1040) or OS (cut-off H-score of ≥190, maximum χ2 of 2.320, P=0.1277). Association between outcome and combined expression of FRA and EGFR We calculated the survival times for patients whose tumors expressed low levels of the receptors FRA and EGFR, high levels of both the receptors or low levels of one receptor and high levels of the other . Patients with tumors expressing low EGFR (H-score < 100) and high FRA levels (H-Score ≥62.5) had the best OS of 16.7 months. 12

ACCEPTED MANUSCRIPT

Discussion In the last decade, several targeted therapies for the treatment of NSCLC were approved and have led to increased survival times of patients with advanced disease. Targeted drugs for the treatment of

RI PT

NSCLC are either monoclonal antibodies (e.g. bevacizumab that is directed against the vascular endothelial growth factor), small molecular TKIs (e.g. EGFR TKIs such as erlotinib, gefitinib, or afatinib or the ALK inhibitor crizotinib) or antifolates such as pemetrexed that is more effective in non-

SC

squamous NSCLC3). Several novel agents targeting FRA are in clinical development for the treatment of NSCLC, and conjugates of a cytotoxic agent with folic acid (e.g. vintafolide)42 or cytotoxic

M AN U

antibodies (e.g. farletuzumab)15,16 are currently being investigated in phase II clinical trials. Therefore, we wanted to study the expression of the EGFR and FRA together. We found that 29% of the patients had high expression of both the receptors and could be candidates for combined targeted therapy. For the detection of EGFR by IHC, several different antibodies are available, but the vast majority of these antibodies detects the external domain of the EGFR. However, amino-terminal

TE D

truncations of the extracellular domain are reported43,44 and the truncated forms retain their function during therapy. Only an antibody detecting specifically the ID (clone 5B7) would also detect truncated

EP

forms of the EGFR that are constitutively active and are inhibited by EGFR TKIs21. Of note, due to steric hindrance, the antibody 5B7 binds only to the active form of the EGFR. Evaluation of EGFR

AC C

protein expression using this antibody for IHC predicted response and outcome to gefitinib in a retrospective study21. In contrast, EGFR assessment by using ED-specific antibodies allows the prediction of clinical benefit from treatment with EGFR-directed antibodies in NSCLC23. However, we wanted to evaluate a rationale for combining EGFR TKIs with FRA-directed agents. Therefore, we decided to use the 5B7 antibody for the present study. FRA and EGFR are each localized in caveolae27,28 and an association between FRA expression and activating EGFR mutations in NSCLC was recently reported13. Because adenocarcinomas of the

13

ACCEPTED MANUSCRIPT lung express higher amounts of FRA13,30,45-48 and patients with pulmonary adenocarcinomas were often treated with pemetrexed, we decided to study the correlation between FRA and EGFR in a cohort of patients receiving pemetrexed-based therapy29. We have previously published data about FRA

RI PT

expression and its correlation to TS and clinicopathological data in a cohort of 207 patients29, but for EGFR protein detection by IHC pre-treatment specimens were available from only 160 patients from the same cohort. However, the results of FRA expression and the clinicopathological data of this

SC

subgroup presented in this work were consistent with the entire group of patients. The comparison of FRA to EGFR expression in this subgroup resulted in a rather similar range of FRA and EGFR

M AN U

expression (Figure 2A and Suppl. Table 1), but FRA expression was not statistically correlated to EGFR expression (Figure 2B).

Pemetrexed is a multitarget antifolate drug approved for the treatment of advanced NSCLC3. While the mechanism for the drug activity is known, only thymidylate synthase (TS) is currently being validated as a predictive biomarker for this agent. In the present study, membranous FRA expression

TE D

was significantly associated with PFS and OS (P=0.0254 and P=0.0076, respectively). Response and clinical disease stabilization were not correlated to membranous expression revealing the question whether FRA might be rather a prognostic biomarker than a predictive biomarker for pemetrexed-based

EP

therapy. Yet, because of a lack of a control group (cohort of patients with pulmonary adenocarcinomas

AC C

without pemetrexed-based treatment) this question cannot be answered in this study. A biomarker subgroup analysis of a phase III clinical trial investigating pemetrexed plus carboplatin versus gemcitabine plus carboplatin examined the association between OS and FRA expression, resulting in a trend to prolonged survival for those with an IHC score in the upper quartile when compared to those in the lowest quartile (10.1 versus 6.3 months; p=0.075). Due to technical reasons, e.g. the antibody used for IHC, Grønberg et al. were not able to accurately distinguish between FRA expression in the cytoplasm and in the membrane49. FRA is biologically active in the membrane only, and the cytoplasmic staining is most likely a combination of some background staining in the 14

ACCEPTED MANUSCRIPT assay and a result of the process of tissue preparation for IHC (as membrane attached proteins, FRA and EGFR are susceptible to leaching or stripping from the membrane during alcohol washes)29. Furthermore, survival analysis was performed in a combined cohort of patients receiving

RI PT

pemetrexed/carboplatin or gemcitabine/carboplatin, putatively masking the association between survival and FRA expression49.

In our study of NSCLC patients with advanced disease, significant differences in FRA

SC

expression between adenocarcinomas and squamous cell carcinomas were found. Our results are consistent with other studies in which FRA was predominantly expressed in adenocarcinomas, relative

M AN U

to squamous cell carcinomas13,30,45-48. However, it needs to be noted that the numbers of patients with “pure” large cell or squamous cell carcinomas were quite low, thereby limiting the significance of our results. Similarly, the number of patients with proven EGFR- or K-Ras mutations (10% each) was lower than expected50. Unfortunately, due to the lack of remaining tissue specimens the EGFR or KRas mutational status is unknown in 28 (18%) or 35 (22%) patients, respectively. It is well known that

TE D

DNA fragmentation is strongly associated with the storage time and that DNA extraction from over 5year-old FFPE blocks yields often a degraded DNA smear33, which hampers the detection of mutations. Some tissue samples used for this study have been stored for an extended period of time, limiting the

EP

use of these samples for DNA analyses. Furthermore, macrodissection of the tumor tissue and direct

AC C

sequencing were performed, but both techniques have a lower sensitivity compared to novel techniques51. Furthermore, it should be mentioned that long-term archived tumor blocks might affect the quality of IHC and that expression analyses of the FRA or EGFR gene were not performed. However, gene expression analyses usually cannot distinguish between functionally active (i.e. membranous) or inactive (i.e. cytoplasmic) receptor levels. Finally, there exist some other limitations. This is a retrospective study, thus only patients with available pre-treatment tumor specimens were included in our study. The clinical response and survival data are based on a retrospective analysis of medical records and image readings and are not collected 15

ACCEPTED MANUSCRIPT from a prospective clinical trial. We collected data from patients receiving pemetrexed-based chemotherapy at different treatment lines and we performed biomarker analyses for this rather heterogeneous group. Several retrospective studies investigating thymidylate synthase as biomarker for

RI PT

pemetrexed-containing chemotherapy have lumped together first-line and further line therapy patients29,52-54. The results of these retrospective analyses have later been validated by prospective studies49,55.

SC

Finally, we performed maximum chi-square statistics to find the optimum cut-off definition resulting in an H-score of 20, which we used to define positive cases. Therefore, our analysis may be prone to type

M AN U

I error similarly to multiple testing bias, and our observations need to be confirmed in an independent series of patients. Other retrospective studies exploring FRA in NSCLC, breast cancer or other gynecologic tumors have used similar low cut-off values (e.g, M-scores of 5 or 10 that are equivalent to H-scores of 30 or 60)30,45,56-58, or a score of 1 on a scale ranging from 0 to 346,59-61 or even any

TE D

membranous expression was considered as positive13.

Conclusion

EP

In advanced NSCLC patients FRA and EGFR are frequently coexpressed, but the FRA and EGFR

AC C

protein levels did not statistically correlate to each other. However, high expression of both the receptors was noted in 29% of the patients.

16

ACCEPTED MANUSCRIPT

Clinical Practice Points (248 words) • In the last decade, targeted therapies such as Tyrosine Kinase Inhibitors (TKIs) of the Epidermal Growth Factor Receptor (EGFR) were approved for the Non-Small Cell Lung

RI PT

Cancer (NSCLC) treatment. The therapeutic antibody farletuzumab targets the Folate Receptor Alpha (FRA) and is currently being investigated in a clinical trial in combination with cisplatin and pemetrexed.

SC

• The interest in combinations of targeted therapies is increasing, but data on the combined

M AN U

expression of the targets is very limited.

• To address this problem, we performed a retrospective study and evaluated the expression of EGFR and FRA in a cohort of 160 advanced NSCLC patients, who underwent pemetrexedbased chemotherapy. We confirmed that FRA is mainly expressed at high levels in pulmonary adenocarcinomas, and that patients whose tumors expressed high levels of FRA in the

TE D

membrane had an improved progression-free and overall survival after chemotherapy with pemetrexed.

EP

• Although in vitro studies reported the detection of both FRA and EGFR in caveloae, which are specialized invaginations of cell membranes, no correlation between EGFR and FRA in

AC C

advanced NSCLC was found. However, both of the receptors were expressed at high levels in the tumor cell membranes in 29% of the patients. This subgroup could potentially be suitable for combined EGFR and FRA-targeted therapies. • Future clinical trials might focus on patients whose tumors strongly express both of the receptors, if a combination of an antifolate, an EGFR TKI, and a therapeutic FRA antibody or an immunotoxin is planned to be administered simultaneously.

17

ACCEPTED MANUSCRIPT

Acknowledgements This work was supported by an IASLC Fellowship Award (D.C.C.) and a research grant from Morphotek, Inc., a subsidiary of Eisai Corporation of North America. The authors would like to

RI PT

acknowledge A. Peglow, S. Fox, H. Loewendick, I. Perelmuter, and R. Daniels for collecting and

AC C

EP

TE D

M AN U

SC

preparing clinical data of the patients, and B. Kamen for scientific advises and critical discussions.

18

ACCEPTED MANUSCRIPT

References 1. Jemal A, Thun MJ, Ries LAG et al. Annual report to the nation on the status of cancer, 19752005, featuring trends in lung cancer, tobacco use, and tobacco control. J Natl Cancer Inst. 2008; 100:1672-1694.

RI PT

2. Peters S, Adjei AA, Gridelli C et al. Metastatic non-small-cell lung cancer (NSCLC): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2012; 23:vii56vii64.

SC

3. Scagliotti GV, Parikh P, von Pawel J et al. Phase III study comparing cisplatin plus gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-smallcell lung cancer. J Clin Oncol 2008; 26:3543-3551.

M AN U

4. Gronberg BH, Bremnes RM, Flotten O et al. Phase III study by the Norwegian lung cancer study group: pemetrexed plus carboplatin compared with gemcitabine plus carboplatin as firstline chemotherapy in advanced non-small-cell lung cancer. J Clin Oncol 2009; 27:3217-3224. 5. Belani CP, Wu Y-L, Chen Y-M et al. Efficacy and safety of pemetrexed maintenance therapy versus best supportive care in patients from east asia with advanced, nonsquamous non-small cell lung cancer: an exploratory subgroup analysis of a global, randomized, phase 3 clinical trial. J Thorac Oncol 2012; 7:567-573.

TE D

6. Ciuleanu T, Brodowicz T, Zielinski C et al. Maintenance pemetrexed plus best supportive care versus placebo plus best supportive care for non-small-cell lung cancer: a randomised, doubleblind, phase 3 study. Lancet 2009; 374:1432-1440. 7. Bearz A, Garassino I, Cavina R et al. Pemetrexed single agent in previously treated non-small cell lung cancer: a multi-institutional observational study. Lung Cancer 2008; 60:240-245. 8. Assaraf YG. Molecular basis of antifolate resistance. Cancer Metastasis Rev 2007; 26:153-181.

EP

9. Zhao R and Goldman ID. Resistance to antifolates. Oncogene 2003; 22:7431-7457.

AC C

10. Chattopadhyay S, Moran RG, and Goldman ID. Pemetrexed: biochemical and cellular pharmacology, mechanisms, and clinical applications. Mol Cancer Ther 2007; 6:404-417. 11. O'Shannessy DJ, Somers EB, Albone E et al. Characterization of the human folate receptor alpha via novel antibody-based probes. Oncotarget 2011; 2:1227-1243. 12. Gruner BA and Weitman SD. The folate receptor as a potential therapeutic anticancer target. Invest New Drugs 1999; 16:205-219. 13. Nunez MI, Behrens C, Woods DM et al. High expression of folate receptor alpha in lung cancer correlates with adenocarcinoma histology and EGFR mutation. J Thorac Oncol 2012; 7:833840. 14. Pribble B and Edelman MJ. EC145: a novel targeted agent for adenocarcinoma of the lung. Expert Opin Investig Drugs 2012; 21:755-761.

19

ACCEPTED MANUSCRIPT 15. Ebel W, Routhier EL, Foley B et al. Preclinical evaluation of MORAb-003, a humanized monoclonal antibody antagonizing folate receptor-alpha. Cancer Immunity 2007; 7:1-8. 16. Thomas A, Maltzman J, and Hassan R. Farletuzumab in lung cancer. Lung Cancer 2013; 80:1518.

RI PT

17. Mendelsohn J. Targeting the epidermal growth factor receptor for cancer therapy. J Clin Oncol 2002; 20:1S-13S. 18. Kim ES, Hirsh V, Mok T et al. Gefitinib versus docetaxel in previously treated non-small-cell lung cancer (INTEREST): a randomised phase III trial. Lancet 2008; 372:1809-1818.

SC

19. Shepherd FA, Rodriques Pereira C, Ciuleanu T et al. Erlotinib in previously treated non-smallcell lung cancer. N Engl J Med 2005; 353:123-132. 20. Hirsch FR, Jänne PA, Eberhardt WE et al. Epidermal growth factor receptor inhibition in lung cancer: status 2012. J Thorac Oncol 2013; 8:373-384.

M AN U

21. Mascaux C, Wynes MW, Kato Y et al. EGFR protein expression in non-small cell lung cancer predicts response to an EGFR tyrosine kinase inhibitor--a novel antibody for immunohistochemistry or AQUA technology. Clin Cancer Res 2011; 17:7796-7807. 22. Pirker R, Pereira JR, Szczesna A et al. Cetuximab plus chemotherapy in patients with advanced non-small-cell lung cancer (FLEX): an open-label randomised phase III trial. Lancet 2009; 373:1525-1531.

TE D

23. Pirker R, Pereira JR, van Pawel J et al. EGFR expression as a predictor of survival for first-line chemotherapy plus cetuximab in patients with advanced non-small-cell lung cancer: analysis of data from the phase 3 FLEX study. Lancet Oncol 2012; 13:33-42.

EP

24. Zappa F, Droege C, Betticher D et al. Bevacizumab and erlotinib (BE) first-line therapy in advanced non-squamous non-small-cell lung cancer (NSCLC) (stage IIIB/IV) followed by platinum-based chemotherapy (CT) at disease progression: a multicenter phase II trial (SAKK 19/05). Lung Cancer 2012; 78:244.

AC C

25. Riggs H, Jalal SI, Baghdadi TA et al. Erlotinib and bevacizumab in newly diagnosed performance status 2 or elderly patients with nonsquamous non-small-cell lung cancer, a phase II study of the Hoosier Oncology Group: LUN04-77. Clin Lung Cancer 2013; 14:224-229. 26. Spigel DR, Edelman MJ, Mok T et al. Treatment rationale study design for the MetLung trial: a randomized, double-blind phase III study of onartuzumab (MetMAb) in combination with erlotinib versus erlotinib alone in patients who have received standard chemotherapy for stage IIIB or IV met-positive non-small-cell lung cancer. Clin Lung Cancer 2012; 13:500-504. 27. Anderson RGW. The caveolae membrane system. Annu Rev Biochem 1998; 67:199-225. 28. Carpenter G. The EGF receptor: a nexus for trafficking and signaling. BioEssays 2000; 22:697707.

20

ACCEPTED MANUSCRIPT 29. Christoph DC, Asuncion BR, Hassan B et al. Significance of folate receptor alpha and thymidylate synthase protein expression in patients with non-small-cell lung cancer treated with pemetrexed. J Thorac Oncol 2013; 8:19-30. 30. O'Shannessy DJ, Yu G, Smale R et al. Folate receptor alpha expression in lung cancer: diagnostic and prognostic significance. Oncotarget 2012.

RI PT

31. Detterbeck FC, Boffa DJ, and Tanoue LT. The new lung cancer staging system. Chest 2009; 136:260-271. 32. Eisenhauer EA, Therasse P, Bogaerts J et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45:228-247.

SC

33. Ludyga N, Grünwald B, Azimzadeh O et al. Nucleic acids from long-term preserved FFPE tissues are suitable for downstream analyses. Virchows Arch 2012; 460:131-140.

M AN U

34. Do H, Krypuy M, Mitchell PL et al. High resolution melting analysis for rapid and sensitive EGFR and KRAS mutation detection in formalin fixed paraffin embedded biopsies. BMC Cancer 2008; 8:doi: 10.1186/1471-2407-8-142. 35. Krypuy M, Newnham GM, Thomas DM et al. High resolution melting analysis for the rapid and sensitive detection of mutations in clinical samples: KRAS codon 12 and 13 mutations in non-small cell lung cancer. BMC Cancer 2006; 6:doi: 10.1186/1471-2407-6-295. 36. Elnakat H and Ratnam M. Distribution, functionality and gene regulation of folate receptor isoforms: implications in targeted therapy. Adv Drug Deliv Rev 2004; 56:1067-1084.

TE D

37. Parker N, Turk MJ, Westrick E et al. Folate receptor expression in carcinomas and normal tissues determined by a quantitative radioligand binding assay. Anal Biochem 2005; 338:284293.

EP

38. Spearman C. The proof and measurement of association between two things. Int J Epidemiol 2010; 39:1137-1150.

AC C

39. Kruskal WH and Wallis WA. Use of ranks in one-criterion variance analysis. Am Stat 1952; 47:583-621. 40. Halpern J. Maximally selected chi square statistics for small samples. Biometrics 1982; 38:1017-1023. 41. Miller R and Siegmund D. Maximally selected chi square statistics. Biometrics 1982; 38:10111016. 42. Pribble B and Edelman MJ. EC145 : a novel targeted agent for adenocarcinoma of the lung. Expert Opin Investig Drugs 2012; 21:755-761. 43. Batra SK, Castelino-Prabhu S, Wikstrand CJ et al. Epidermal growth factor ligand-independent, unregulated, cell-transforming potential of a naturally occurring human mutant EGFRvIII gene. Cell Growth Differ 1995; 6:1251-1259.

21

ACCEPTED MANUSCRIPT 44. Sok JC, Coppelli FM, Thomas SM et al. Mutant epidermal growth factor receptor (EGFRvIII) contributes to head and neck cancer growth and resistance to EGFR targeting. Clin Cancer Res 2006; 12:5064-5073.

RI PT

45. Bremer RE, Scoggin TS, Somers EB et al. Interobserver agreement and assay reproducibility of folate receptor a expression in lung adenocarcinoma: a prognostic marker and potential therapeutic target. Arch Pathol Lab Med 2012; 137:1747-1752. 46. Cagle PT, Zhai QJ, Murphy L et al. Folate receptor in adenocarcinoma and squamous cell carcinoma of the lung: potential target for folate-linked therapeutic agents. Arch Pathol Lab Med 2012; 137:241-244.

SC

47. Franklin WA, Waintrub M, Edwards D et al. New anti-lung-cancer antibody cluster 12 reacts with human folate receptors present on adenocarcinoma. Int J Cancer 1994; 57:89-95. 48. Iwakiri S, Sonobe M, Nagai S et al. Expression status of folate receptor alpha is significantly correlated with prognosis in non-small-cell lung cancers. Ann Surg Oncol 2008; 15:889-899.

M AN U

49. Gronberg BH, Lund-Iversen M, Strom EH et al. Associations between TS, TTF-1, FR-alpha, FPGS, and overall survival in patients with advanced non-small cell lung cancer receiving pemetrexed plus carboplatin or gemcitabine plus carboplatin as first-line chemotherapy. J Thorac Oncol 2013; 8:1255-1264. 50. Pao W, Iafrate AJ, and Su Z. Genetically informed lung cancer medicine. J Pathol 2011; 223:230-240.

TE D

51. Borras E, Jurado I, Hernan I et al. Clinical pharmacogenomic testing of KRAS, BRAF and EGFR mutations by high resolution melting analysis and ultra-deep pyrosequencing. BMC Cancer 2011; 11.

EP

52. Chen C-Y, Chang Y-L, Shih J-Y et al. Thymidylate synthase and dihydrofolate reductase expression in non-small cell lung carcinoma: the association with treatment efficacy of pemetrexed. Lung Cancer 2011; 74:132-138.

AC C

53. Sun J-M, Han J, and Ahn JS. Significance of thymidylate synthase and thyroid transcription factor 1 expression in patients with nonsquamous non-small cell lung cancer treated with pemetrexed-based chemotherapy. J Thorac Oncol 2011; 6:1392-1399. 54. Igawa S, Ryuge S, Wada M et al. Pemetrexed for previously treated patients with non-small cell lung cancer and differences in efficacy according to thymidylate synthase expression. Chemotherapy 2012; 58:313-320. 55. Nicolson MC, Fennell Da, Ferry D et al. Thymidylate synthase expression and outcome of patients receiving pemetrexed for advanced nonsquamous non-small-cell lung cancer in a prospective blinded assessment phase II clinical trial. J Thorac Oncol 2013; 8:930-939. 56. O`Shannessy DJ, Somers EB, Maltzman J et al. Folate receptor alpha (FRA) expression in breast cancer: identification of a new molecular subtype and association with triple negative disease. Springerplus 2012; 1.

22

ACCEPTED MANUSCRIPT 57. O`Shannessy DJ, Somers EB, Smale R et al. Expression of folate receptor-a (FRA) in gynecologic malignancies and its relationship to the tumor type. Int J Gynecol Pathol 2012; 32:258-268.

RI PT

58. Despierre E, Lambrechts S, Leunen K et al. Folate receptor alpha (FRA) expression remains unchanged in epithelial ovarian and endometrial cancer after chemotherapy. Gynecol Oncol 2013; 130:192-199. 59. Hartmann LC, Keeney GL, Lingle WL et al. Folate receptor overexpression is associated with poor outcome in breast cancer. Int J Cancer 2007; 121:938-942.

SC

60. Nutt JE, Razak ARA., O'Toole K et al. The role of folate receptor alpha (FRalpha) in the response of malignant pleural mesothelioma to pemetrexed-containing chemotherapy. Br J Cancer 2010; 102:553-560.

AC C

EP

TE D

M AN U

61. Saba NF, Wang X, Müller S et al. Examining expression of folate receptor in squamous cell carcinoma of the head and neck as a target for a novel nanotherapeutic drug. Head Neck 2009; 31:475-481.

23

ACCEPTED MANUSCRIPT

Legend to Figure 1: IHC for Folate Receptor Alpha (FRA) and Epidermal Growth Factor Receptor (EGFR) in Non-Small Cell Lung Cancer (NSCLC) samples with different protein expression levels. A-D: IHC with an

RI PT

antibody against FRA (dilution 2.5 µg/ml). A) renal tissue was used as positive control, B) weakly, C) moderately, and D) strongly stained NSCLC sample. E-H: IHC with an antibody against the intracellular domain of EGFR (dilution ≈0.4 µg/ml). E) NSCLC specimen with known EGFR

SC

expression was used as positive control; F) weakly, G) moderately, and H) strongly stained NSCLC

M AN U

sample. Magnification 1:400.

Legend to Figure 2:

Vertical scatter plot (A) showing distribution of membranous protein levels of Folate Receptor Alpha and Epidermal Growth Factor Receptor (EGFR). Intratumoral correlation (B) of membranous

TE D

expression of FRA or EGFR.

Legend to Figure 3:

EP

Vertical scatter plots of membranous expression of (A) Folate Receptor Alpha (FRA) or (B) Epidermal Growth Factor Receptor (EGFR) in different histological subtypes (horizontal line: median). ADC:

AC C

adenocarcinoma; LCC: large cell carcinoma; SCC: squamous cell carcinoma.

Legend to Figure 4:

Analyses of expression levels of Folate Receptor Alpha (FRA) and (A) progression-free survival (PFS) or (B) overall survival in the entire cohort. Subgroup analyses of patients with first-line therapy only (C and D) or of patients receiving second-line therapy only (E and F).

24

ACCEPTED MANUSCRIPT

Table 1 Epidemiologic and clinicopathological data

59 (23-83)

RI PT

74 (46%) 86 (54%)

SC

22 (14%) 64 (40%) 23 (14%) 51 (32%)

AC C

EP

28 (18%) 13 (8%) 64 (40%) 55 (34%)

M AN U

TE D

Age Median age at initial diagnosis in years (range) Gender Female Male T-stage (at initial diagnosis) T1 T2 T3 T4 N-stage (at initial diagnosis) N0 N1 N2 N3 M-stage (at initial diagnosis) M0 M1 IASLC-stage (at initial diagnosis) I II III IV Histologic subtype “pure” adenocarcinoma “pure” large cell carcinoma “pure” squamous cell carcinoma mixed and other carcinomas Histological grading G1 G2 G3 G4 Unspecified

Number of patients (%) 160 (100%)

54 (34%) 106 (66%) 6 (4%) 8 (5%) 40 (25%) 106 (66%)

123 (77%) 9 (6%) 8 (5%) 20 (13%) 7 (4%) 49 (31%) 100 (63%) 1 (<1%) 3 (2%)

ACCEPTED MANUSCRIPT

Table 1 (continued) Number of patients (%) 160 (100%)

RI PT

51 (32%) 27 (17%) 1 (<1%) 4 (3%) 77 (48%) 4 (1-18) 3,470 (600-18,000) 2,9 (0.7-14.6) 54 (34%) 51 (32%) 30 (19%) 25 (16%)

2 (1%) 58 (37%) 57 (36%) 30 (19%) 13 (8%) 4.9 (4.1-5.7) 10.0 (7.7-12.4)

EP

Legend to Table 1:

TE D

M AN U

Chemotherapy Pemetrexed plus cisplatin Pemetrexed plus carboplatin Pemetrexed plus cisplatin and carboplatin Pemetrexed plus other compounds Pemetrexed single-agent Median number of cycles (range) Median cumulative pemetrexed dose in mg (range) Median duration of pemetrexed-based therapy in months (range) Chemotherapy line 1st line 2nd line 3rd line ≥4th line Response to pemetrexed-based treatment CR PR SD PD Not evaluable Survival time from start of treatment to progression or death Median PFS time in months (95% CI) Median OS time in months (95% CI)

SC

Characteristics

AC C

Clinicopathological data of NSCLC patients. TNM and IASLC stages refer to the initial diagnosis of NSCLC. CR: complete remission; IASLC: International Association for the Study of Lung Cancer; OS: overall survival; PD: progressive disease; PFS: progression-free survival; PR: partial remission; and SD: stable disease. Note that percentages may not total 100 due to rounding off.

ACCEPTED MANUSCRIPT

Table 2A Number of patients (%) 74 (46%) 86 (54%)

93 ± 9, 80 71 ± 7, 45

123 (77%) 9 (6%) 8 (5%) 20 (13%)

86 ± 7, 80 36 ± 13, 35 44 ± 29, 0 85 ± 16, 60

4 (3%) 36 (28%) 86 (66%) 2 (2%)

61 ± 23, 60 91 ± 10, 100 80 ± 8, 60 0 ± 0, 0

RI PT

PKW=0.0328

22 (14%) 64 (40%) 23 (14%) 51 (32%) 28 (18%) 13 (8%) 64 (40%) 55 (34%)

SC

PKW=0.2719

PKW=0.5248

97 ± 13, 112.5 85 ± 9, 80 70 ± 14, 65 74 ± 11, 45

PKW=0.8563

83 ± 12, 77.5 87 ± 16, 90 80 ± 10, 60 80 ± 10, 60

54 (34%) 106 (66%)

85 ± 10, 70 79 ± 7, 62.5

6 (4%) 8 (5%) 40 (25%) 106 (66%)

58 ± 29, 35 76 ± 18, 85 91 ± 13, 70 79 ± 7, 62.5

119 (90%) 10 (10%)

82 ± 7, 65 100 ± 23, 80

P=0.5355

112 (90%) 13 (10%)

89 ± 7, 80 69 ± 18, 40

P=0.4274

EP

AC C

P-value P=0.0769

TE D

Gender Female Male Histological subtype “pure” adenocarcinoma “pure” large cell carcinoma “pure” squamous cell carcinoma mixed and other carcinomas Grading G1 G2 G3 G4 T-stage at initial diagnosis T1 T2 T3 T4 N-stage at initial diagnosis N0 N1 N2 N3 M-stage at initial diagnosis M0 M1 IASLC-Stage at initial diagnosis IASLC I IASLC II IASLC III IASLC IV EGFR status wild-type mutant K-Ras status wild-type mutant

Mean membranous FRA H-score ± SEM, median

M AN U

Characteristic

P=0.6576 PKW=0.7410

ACCEPTED MANUSCRIPT

Table 2B Number of patients (%) 74 (46%) 86 (54%)

92 ± 7, 97.5 97 ± 7, 100

123 (77%) 9 (6%) 8 (5%) 20 (13%)

93 ± 6, 100 88 ± 17, 80 115 ± 23, 127.5

4 (3%) 36 (28%) 86 (66%) 2 (2%)

87 ± 15, 100 82 ± 8, 95 102 ± 7, 102.5 75 ± 0, 75

RI PT

PKW=0.7251

22 (14%) 64 (40%) 23 (14%) 51 (32%) 28 (18%) 13 (8%) 64 (40%) 55 (34%)

SC

PKW=0.4909

PKW=0.8874

97 ± 15, 102.5 94 ± 8, 95 103 ± 12, 110 91 ± 8, 100

PKW=0.0722

115 ± 12, 110 93 ± 22, 80 78 ± 7, 92.5 105 ± 9, 105

54 (34%) 106 (66%)

82 ± 9, 80 101 ± 6, 100

6 (4%) 8 (5%) 40 (25%) 106 (66%)

71 ± 30, 60 88 ± 22, 87.5 83 ± 10, 95 101 ± 6, 100

119 (90%) 10 (10%)

94 ± 6, 100 126 ± 15, 120

P=0.0750

112 (90%) 13 (10%)

98 ± 6, 102.5 104 ± 16, 80

P=0.8239

EP

AC C

P-value P=0.6134

TE D

Gender Female Male Histological subtype “pure” adenocarcinoma “pure” large cell carcinoma “pure” squamous cell carcinoma mixed and other carcinomas Grading G1 G2 G3 G4 T-stage at initial diagnosis T1 T2 T3 T4 N-stage at initial diagnosis N0 N1 N2 N3 M-stage at initial diagnosis M0 M1 IASLC-Stage at initial diagnosis IASLC I IASLC II IASLC III IASLC IV EGFR status wild-type mutant K-Ras status wild-type mutant

Mean membranous EGFR H-score ± SEM, median

M AN U

Characteristic

P=0.0820 PKW=0.3148

ACCEPTED MANUSCRIPT

Legend to Table 2: Associations between clinicopathological data and membranous expression of (A) Folate Receptor Alpha (FRA) and or (B) Epidermal Growth Factor Receptor (EGFR). TNM and IASLC stages refer to

RI PT

the initial diagnosis of NSCLC. IASLC, International Association for the Study of Lung Cancer; KRas; Kirsten RNA-associated rat Sarcoma 2 virus. SEM, standard error of the mean. Note that

AC C

EP

TE D

M AN U

SC

percentages may not total 100 due to rounding off.

B

E

F

C

D

AC C

EP

TE D

M AN U

SC

A

RI PT

ACCEPTED MANUSCRIPT

G

H

ACCEPTED MANUSCRIPT

A

distribution of membranous protein levels

RI PT

200

B

correlation of membranous expression r=0.1136 P=0.1527

EP

300

200

AC C

H-score EGFR

EG

TE D

FR A

0

FR

SC

100

M AN U

H-score

300

100

0 0

100

200

H-score FRA

300

ACCEPTED MANUSCRIPT

200

100

M AN U

100

200

SC

EGFR H-score

300

0

EP

TE D

SC C

LC C

D C

SC C

AC C

LC C

D C

0

A

FRA H-score

300

EGFR membranous expression

RI PT

B

FRA membranous expression

A

A

ACCEPTED MANUSCRIPT

80

median PFS: 3.4 months (n=48) median PFS: 5.5 months (n=112)

60

P=0.0254 HR=0.6060 (95%CI: 0.3906-0.9403)

40 20

10

20

30

40

20

D

median OS: 6.4 months (n=48) median OS: 12.1 months (n=112) P=0.0076 HR=0.5726 (95%CI: 0.3803-0.8621)

median PFS: 3.7 months (n=18) median PFS: 7.8 months (n=36)

60

20

TE D

P<0.0001 HR=0.1471 (95%CI: 0.0575-0.3761)

40

0 0

10

20

30

40

AC C

membranous FRA expression 100

FRA H-score <20 FRA H-score >=20

80

median PFS: 2.8 months (n=16) median PFS: 4.6 months (n=35)

60

P=0.1481 HR=0.5386 (95%CI: 0.2329-2.177)

40 20 0

40

60

80

100

OS in months

membranous FRA expression FRA H-score <20 FRA H-score >=20

80

median OS: 7.9 months (n=18) median OS: 22.2 months (n=36)

60

P=0.0060 HR=0.3144 (95%CI: 0.1377-0.7145)

40 20 0 0

20

40

60

80

OS in months

EP

PFS in months

20

100

Percent survival

FRA H-score <20 FRA H-score >=20

80

M AN U

membranous FRA expression 100

Percent survival

40

0

PFS in months

Percent survival

60

0

0

E

FRA H-score <20 FRA H-score >=20

80

SC

0

C

100

Percent survival

FRA H-score <20 FRA H-score >=20

membranous FRA expression

F

100

Percent survival

Percent survival

100

membranous FRA expression

B

RI PT

membranous FRA expression

A

FRA H-score <20 FRA H-score >=20

80

median OS: 4.6 months (n=16) median OS: 11.6 months (n=35)

60

P=0.0273 HR=0.4430 (95%CI: 0.2149-0.9129)

40 20 0

0

10

20

PFS in months

30

0

20

40

60

OS in months

80

100

ACCEPTED MANUSCRIPT

Suppl. Table 1

EGFR (H-score for membranous expression) 0 40 100 130 240 95 62 5

RI PT

minimum 25% percentile median 75% percentile maximum mean SD SEM

FRA (H-score for membranous expression) 0 10 62.5 140 275 81 72 6

SC

PROTEIN

M AN U

Legend to Suppl. Table 1:

Distribution of membranous protein expression of Folate Receptor Alpha (FRA) or Epidermal Growth

AC C

EP

TE D

Factor Receptor (EGFR). SD: standard deviation; SEM: standard error of the mean.

ACCEPTED MANUSCRIPT

Suppl. Table 2

membranous FRA expression < median H-score of 62.5 41 patients (26%) 39 patients (24%)

≥ median H-score 62.5 33 patients (21%) 47 patients (29%)

RI PT

membranous EGFR expression

< median H-score of 100 ≥ median H-score of 100

Legend to Suppl. Table 2:

Distribution of patients with tumors expressing low levels of both Folate Receptor Alpha (FRA) or

SC

Epidermal Growth Factor Receptor (EGFR), low FRA levels and high EGFR levels, high FRA and low EGFR levels, or high levels of both the receptors. Assessment of membranous protein levels was

AC C

EP

TE D

M AN U

performed by the H-score.

ACCEPTED MANUSCRIPT

Suppl. Table 3

Number of patients Mean FRA H-score ± SEM, median 87 ± 9, 80 82 ± 8, 60

RI PT

63 87

P-value 0.6286

0.4613

117 13

86 ± 7, 80 78 ± 15, 42.5

SC

3A OR PR SD + PD DC PR + SD PD

Number of patients Mean FRA H-score ± SEM, median

3C OR PR SD + PD DC PR + SD PD

Number of patients Mean FRA H-score ± SEM, median

67 ± 13, 40 79 ± 15, 45

47 3

76 ± 11, 40 30 ± 15, 40

TE D

27 23

P-value 0.4584

0.4128

P-value 0.4890

104 ± 18, 115 86 ± 15, 65

0.6931

EP

15 31

88 ± 12, 95 101 ± 26, 85

AC C

32 14

M AN U

3B OR PR SD + PD DC PR + SD PD

Legend to Suppl. Table 3:

Associations between protein expression of FRA and objective response (OR) or disease control (DC). Responding and non-responding patients were compared, but no significant differences in the mean or median FRA H-Score for OR or DC were found for the entire cohort (3A), for patients with first-line chemotherapy only (3B) or for patients with second-line chemotherapy only (3C).