Molecular in vitro diagnostic test for the quantitative detection of the mrna expression of ERBB2, ESR1, PGR and MKI67 in breast cancer tissue.

Innovation for your breast cancer diagnostics Now valid ated on: -Roche cob as z 480 A nalyzer -Roche L ightcycle r 480 II - ABI 750 0 Fast -Versant kpcr Cyc ler PGR ESR1 MKI67 Molecular in vitro diagnostic test for the quantitative detection of the mrna expression of ERBB2, ESR1, PGR and MKI67 in breast cancer tissue. G ATA G C G A C G AT C G A A A G A A G T TA G ATA G C G A C G AT C G A A A G A A G T TA G ATA G C G A C G AT C G A A A G A A G T TA G ATA G C G A C ERBB2

Medical Need European Guidelines 1 recommend that all breast cancer tumors be routinely subtyped by using a combination of four molecular markers ERBB2 (HER2), ESR1 (ER), PGR (PR) and MKI67 (marker of proliferation Ki-67) It has become evident that different molecular subtypes of breast cancer tumors require a different treatment approach. Thus a precise separation of the different subtypes is one key parameter to define a successful treatment strategy Innovative RT- qpcr technologies open up new dimensions in breast cancer diagnostics RT-qPCR has a series of widely acknowledged methodological advantages, it is - quantitative by nature with a wide dynamic range 2 - precise - highly sensitive - objective - reproducible and reliable 2

Fixing today s obstacles in breast cancer biomarker assessments and subtyping Key Characteristics: 1. High performing test Molecular in vitro diagnostic test for four essential breast cancer biomarkers ERBB2 (HER2), ESR1 (ER), PGR (PR) and MKI67 (marker of proliferation Ki-67) Reproducible, quantitative determination of mrna expression of each marker via RT-qPCR 3 2. Promising clinical utility Results enable precise and accurate molecular subtyping of tumor tissue acc. to St Gallen into Luminal A-like, (HER2 negative), (HER2 positive), HER2 positive (non-luminal) and Triple negative Validated in retrospective clinical performance evaluation studies analyzing more than 800 patient samples 4,5 Extended information for a patient s prognosis and treatment selection based on subtyping 3. Easy to use Stable, reliable and reproducible process for any molecular pathology laboratory From biopsy FFPE sample to result in 1 day Accurate molecular subtyping as a basis for clinical decision making Benefit Precise quantitative measurement of biomarker expression is a prerequisite for future refinement of breast cancer diagnostics and treatment 3

MammaTyper delivers precise quantitative results for each biomarker 1. MammaTyper result: Quantitative values (- Cq) per marker 2. Each marker is classified as positive or negative according to validated cut-off values: ERBB2 Negative < 41.10 Cut-off 41.10 Positive 41.10 <26 26 27 28 29 31 32 33 34 36 37 38 39 41 42 43 44 46 >46 Negative < 38.00 Cut-off 38.00 Positive 38.00 ESR1 <26 26 27 28 29 31 32 33 34 36 37 38 39 41 42 43 44 46 >46 Negative <.50 Cut-off.50 Positive.50 PGR <26 26 27 28 29 31 32 33 34 36 37 38 39 41 42 43 44 46 >46 Negative <.90 Cut-off.90 Positive.90 MKI67 <26 26 27 28 29 31 32 33 34 36 37 38 39 41 42 43 44 46 >46 Values in the ranges coloured in light blue/orange have been observed in a cohort of 765 breast cancer samples. Values in the ranges coloured in dark blue/orange have not yet been observed (clinical thresholds shown for Roche LightCycler 480 instrument II; Cut-offs validated in the FinHer trial with 765 breast cancer samples, for study details see page 9). 3. The ΔCq value provides a quantitative measure for the degree of overexpression of each marker (vs. the cut-off) - Cq values with corresponding linear fold changes relative to cutoff Observed expression value with standard deviation Linear fold change relative to cut-off 20 15 10 5 0 MKI67 <32 32 33 34 36 37 38 39 > - Cq Example: -Δ Cq value 39.00 positive 11.3x above cut-off Cut-off Upper/lower border of range of typically observed - Cq values The logarithmic - Cq scale is particularly well suited to depict the high resolution and wide range of values for marker expression that can be observed. Standard deviations depicted here were determined in a three site technical validation study. For detailed information please refer to the instructions for use. MammaTyper delivers precise and reproducible - Cq values, which are the basis for further differentiation of subtypes beyond today s routine results an important factor for more treatment individualization in the future. 4

Quantitative biomarker assessments lead to accurate subtyping The combination of marker results reveal the St Gallen subtypes: Luminal A-like, (HER2 negative), (HER2 positive), HER2 positive (non-luminal) and Triple negative tumors Prognosis and therapy decision is based on the subtype and other clinico-pathological factors ERBB2 ESR1 PGR MKI67 St Gallen Subtype 1 ERBB2 ESR1 PGR MKI67 St Gallen Subtype 1 pos pos pos pos pos pos pos neg pos pos neg pos pos pos neg neg (HER2 positive) (HER2 positive) (HER2 positive) (HER2 positive) neg pos pos pos (HER2 negative) neg pos pos neg Luminal A-like neg pos neg pos neg pos neg neg (HER2 negative) (HER2 negative) pos neg pos pos Not defined 1 neg neg pos pos Not defined 1 pos neg pos neg Not defined 1 neg neg pos neg Not defined 1 pos neg neg pos HER2 positive (non-luminal) neg neg neg pos Triple negative (ductal) pos neg neg neg HER2 positive (non-luminal) neg neg neg neg Triple negative (ductal) The not defined subtypes according to St Gallen (2013) are based on the assumption that a combination showing an ER negative but a PR positive result is not possible. In an epidemiological inquiry 2.4% ER negative/pr positive tumors have been found in an American breast cancer cohort (Ries et al. 2007). These subtypes were also found in two MammaTyper clinical performance evaluation studies in some cases (3.5%). 5

MammaTyper vs. current standard methods MammaTyper results show a high degree of agreement with IHC results for ERBB2/HER2, ESR1/ER and PGR/PR 4 pos IHC neg Agreement varies by marker as expected: High overall agreement for ERBB2/HER2, ESR1/ER, PGR/PR pos PCR neg 18% 4% ERBB2/HER2 4% 75% K 0,79 p< 0.0001 OPA 93% PPA 83% NPA 95% 70% 5% ESR1/ER 3% 22% K 0,80 p< 0.0001 OPA 92% PPA 96% NPA 81% Lowest overall agreement for MKI67/Ki-67 OPA = Overall percent agreement NPA = Negative percent agreement PPA = Positive percent agreement 54% 13% PGR/PR K 0,64 p< 0.0001 OPA 83% 55% 19% MKI67/Ki-67 K 0, p< 0.0001 OPA 75% Agreement of mrna expression values determined by MammaTyper with protein expression determined by IHC (IHC/CISH for HER2) in the FinHer samples. 4% 29% PPA 93% NPA 69% 5% 21% PPA 91% NPA 51% MKI67 MammaTyper results are more accurate in discordant cases compared to IHC results according to clinical outcome 4 Analysis of discordant cases shows: Better distant disease-free survival could be shown as a trend, if MKI67 determined by MammaTyper was negative and worse if MKI67 determined by MammaTyper was positive in discordant cases (i.e. in agreement with the concordant cases in this study). Fractions of patients surviving 1,0 0,9 0,8 neg pos Survival curves by MKI67 (MammaTyper ) Kaplan-Meier analysis for the MammaTyper results of MKI67 discrepant cases by RT-qPCR and IHC method. 5-year distant disease-free survival rate is shown for IHC ER positive patients (n=143). MammaTyper MKI67 negative 97% DDFS (blue) vs MammaTyper MKI67 positive 86% DDFS (red); p=0.123; HR 0.205; 95% CI=0.027-1.541 (based on the FinHer trial. For more trial information please see page 9). 0,7 0 2 4 6 Time to distant recurrence [years] 6

MammaTyper reproducibility and stability MammaTyper is a highly reproducible test, which avoids disadvantages of intra-/inter-laboratory variation 3 ERBB2 ESR1 Performance characterization: - Cq 1 2 3 4 5 6 7 Ref. 8RNA PGR 1 2 3 4 5 6 7 Ref. 8RNA MKI67 Samples: RNA extracted from FFPE tumor samples of 7 patients and 1 reference RNA Tests: Total of 24 measurements per sample: Each patient RNA was measured 2 times on 3 days, on 4 different LightCycler 480 instruments II On display: - Cq values for the four markers - Cq 1 2 3 4 5 6 7 Ref. 8RNA 1 2 3 4 5 6 7 Ref. 8RNA A marker is evaluated as positive (pos) at or above the cut-off value and negative (neg) below the cut-off value (black dashed line) Test performance is not affected by fluctuations in tumor cell content (TCC). This leads to an agreement of results between enriched and non-enriched samples 3 ERBB2 ESR1 Performance characterization: - Cq 1 2 3 4 5 6 7 8 9 Sample pair number PGR 1 2 3 4 5 6 7 8 9 Sample pair number MKI67 Paired samples with high (80%) and low (5-50%) tumor cell content have been measured Comparison of results of enriched samples and non enriched samples - Cq 1 2 3 4 5 6 7 8 9 Sample pair number 1 2 3 4 5 6 7 8 9 Sample pair number Non enriched ----- Cut-off Enriched -Δ Cq values of paired tumor samples with high (80%, gray) and low (5 50%, green) tumor cell content and varying DCIS (10 70%) content adjacent to the tumor. 7

MammaTyper is an easy to handle test which allows same day results Step1: RNXtract sample preparation Step2: MammaTyper test preparation and execution Cut 10µm section FFPE biopsy FFPE curl RNXtract Total RNA template MammaTyper 1 2 3 4 5 6 7 8 9 10 11 12 A P1 P1 P1 P1 P1 P1 P1 P1 P1 PC PC PC B P2 P2 P2 P2 P2 P2 P2 P2 P2 PC PC PC C P3 P3 P3 P3 P3 P3 P3 P3 P3 PC PC PC D P4 P4 P4 P4 P4 P4 P4 P4 P4 - - - Run RT-qPCR E P5 P5 P5 P5 P5 P5 P5 P5 P5 - - - F P6 P6 P6 P6 P6 P6 P6 P6 P6 NC NC NC G P7 P7 P7 P7 P7 P7 P7 P7 P7 NC NC NC H P8 P8 P8 P8 P8 P8 P8 P8 P8 NC NC NC Data analysis Real Time PCR Instrument Versant kpcr Cycler, Roche LightCycler 480 II, ABI 7500 Fast or cobas z 480 Analyzer Distribute master mixes, 8 RNA templates, PC and NC in 96 well plate Set up 3 RT-qPCR master mixes RT-qPCR less time-consuming compared to IHC FISH. RNXtract Manual steps (hands-on time) Walk away time MammaTyper Conducted for all 4 markers simultaniously Pathologist time 8:00 9:00 10:00 11:00 12:00 13:00 14:00 MammaTyper result for four markers, including result for ERBB2, in 1 day. Sample processing from section to result in only 5: h for up to 8 patients. MammaTyper reduces hands-on time of technician as well as pathologist time 8

Results from clinical performance evaluation studies / hypothesis generating data MammaTyper provides prognostic value through reliable subtyping: MammaTyper results separated low risk Luminal A-like patients accurately from the group of patients with other subtypes with regard to clinical outcome in the FinHer trial 4 Survival curves by Luminal-A-like subgroup (MammaTyper ) Survival curves by Luminal-A-like subgroup (MammaTyper ) Fractions of patients surviving 1,0 0,9 0,8 0,7 Luminal-A-like Combined subgroup Fractions of patients surviving 1,0 0,9 0,8 0,7 Luminal-A-like Combined subgroup 0 2 4 6 Time to distant recurrence [years] 0 2 4 6 Time to death [years] Kaplan Meier analysis of DDFS for patients with up to three positive nodes (n=536). 5-year distant disease-free survival of 97% for Luminal A-like patients versus 88% for other subtypes (p=0.0062; HR 0.241; 95% CI=0.087-0.668) (Combined subgroup = all other subtypes combined). Kaplan Meier analysis of overall survival for patients with up to three positive nodes (n=536). 5-year overall survival of 99% of the Luminal A-like patients versus 94% for other subtypes (p=0.0289; HR 0.108; 95%-CI=0.015-0.795) (Combined subgroup = all other subtypes combined). MammaTyper provides predictive value by identifying subtypes with Taxane benefit by accurate MKI67 determination: Patients classified as (HER2 negative) by MammaTyper showed a significant benefit from Taxane addition in the FinHer trial 4 A significant benefit of docetaxel addition was observed only for the Luminal B like (HER2 negative) subtype, as already published previously for other studies (Nitz et al. 2014). Survival curves by Chemotherapy agent Survival curves by Chemotherapy agent Fractions of patients surviving 1,0 0,9 0,8 0,7 Docetaxel Vinorelbine Fractions of patients surviving 1,0 0,9 0,8 0,7 Docetaxel Vinorelbine 0 2 4 6 0 2 4 6 Time to distant recurrence [years] Time to death [years] Kaplan Meier analysis of distant disease-free survival of (HER2 negative) patients (n=315) treated either with docetaxel (blue) or vinorelbine (red) in addition to FEC. 5-year distant disease-free survival rate was 88% after docetaxel and 78% after vinorelbine treatment (p=0.0196; HR 0.514; 95%-CI=0.294-0.899). Kaplan Meier analysis of overall survival for (HER2 negative) patients (n=315) treated either with docetaxel (blue) or vinorelbine (red) in addition to FEC. 5-year overall survival was 97% after docetaxel and 89% after vinorelbine treatment (p=0.0115; HR 0.6; 95%-CI=0.122-0.767). FinHer Study: Clinical performance of the MammaTyper kit were validated based on a retrospective analysis of samples which have been collected prospectively during the FinHer-trial (International Controlled Trial: ISRCTN76560285). The aims of the study were to determine the prognostic and predictive validity of MammaTyper results for node negative and node positive breast cancer patients receiving adjuvant systemic chemotherapy as well as to compare MammaTyper with IHC results. The study cohort comprised 801 FFPE blocks which could be collected retrospectively from patients of the FinHer-study cohort, recruited between Oct 2000 and Sept 2003 (Joensuu et al. 2006, Joensuu et al 2009). 9

Results from clinical performance evaluation studies / hypothesis generating data MammaTyper helps to identify those patients who are likely to benefit from neoadjuvant chemotherapy based on accurate determination of MKI67: Based on the results of the first MammaTyper neoadjuvant study the diagnostic accuracy of MammaTyper compared to IHC has been seven-times better in predicting the use of neoadjuvant chemotherapy 5 ROC curves for Path. compl. Response 100 10 Sensitivity 1,0 0,8 0,6 Ki-67 vihc [% positive nuclei] 80 60 20 0,4 0 0,2 no-pcr pcr pcr vihc MammaTyper 0,0 0,0 0,2 0,4 0,6 0,8 1,0 10 1 - Specificity ROC analysis for pathological complete response using vihc Ki-67 measurements (blue) and continuous MKI67 values measured with MammaTyper (green). ROC curves show a better identification of responders versus non-responders by mrna determination. MKI67 MammaTyper [-Δ Cq] 39 38 37 36 64 4 no-pcr pcr pcr Box plots showing the distribution of vihc results (above) and continuous RT-qPCR mrna results (below) in relation to the group of non-responders (no-pcr) and responders (pcr) (n=53). The differences were investigated using Mann-Whitney test.. 10

The benefit of RT-qPCR approach for improved subtyping of breast cancer is supported by many publications Selected Literature References / Quotes RT-qPCR versus IHC RT-qPCR has a series of widely acknowledged methodological advantages over IHC, which appear particularly beneficial in the context of reducing the bias of routine Ki-67 assessment; it is quantitative by nature with much wider dynamic range, it does not require an experienced eye and results are not affected by subjective interpretations (Noske, Loibl, Darb-Esfahani et al., Comparison of different approaches for assessment of HER2 expression on protein and mrna level: prediction of chemotherapy response in the neoadjuvant GeparTrio trial. Breast Cancer Res. Treat (2011): 109-17.). Ki-67 prognostic value Ki-67 is an excellent surrogate for proliferation and broadly accepted. With respect to clinical application, we revealed connections and equivalence between traditional prognostic factors, expression-based subtyping, and prognostic signatures and provided evidence that these signatures should be tested for their ability to spare adjuvant chemotherapy mainly in the low-proliferation subgroup of patients with ER tumors. (Wirapati, Sotiriou et al., Meta-analysis of gene expression profiles in breast cancer: toward a unified understanding of breast cancer subtyping and prognosis signatures. Breast Cancer Res. (2008).). Ki-67 prognostic value Ki-67 has predictive and prognostic value and is a feasible marker for clinical practice. It independently improved the prediction of treatment response and prognosis in a group of breast cancer patients receiving neoadjuvant treatment. As mean Ki-67 values in patients with a pcr were very high, cut-off values in a high range above which the prognosis may be better than in patients with lower Ki-67 values may be hypothesized. (Fasching et al., Ki-67, chemotherapy response, and prognosis in breast cancer patients receiving neoadjuvant treatment. BMC Cancer (2011): 486.). Ki-67 prognostic value In patients with a low-proliferation primary tumor, a high level of PR expression would indicate an extraordinarily good prognosis (HR 0.39; 95 % CI, 0.23 0.66). In patients with higher-proliferation primary tumors, PR status was not capable of differentiating prognostic groups. Ki-67 is useful in addition to known prognostic factors for breast cancer. It is able to indicate a group of women with a poorer prognosis, specifically in the group of patients with PR-positive breast cancer. (Loehberg et al., Prognostic relevance of Ki-67 in the primary tumor for survival after a diagnosis of distant metastasis. Breast Cancer Res. Treat. (2013): 899-908.). Ki-67 neoadjuvant therapy Despite methodological concerns, overall a strong correlation of Ki-67 with breast cancer outcome is sufficiently supported, particularly by data originating from randomized clinical trials with central review of biomarkers (Luporsi, Andre, Spyratos et al., Ki-67: level of evidence and methodological considerations for its role in the clinical management of breast cancer: analytical and critical review. Breast Cancer Res. Treat. (2012): 859-915.). This has also been shown in the neoadjuvant setting, whereby higher Ki-67 values are consistently associated with higher rates of pathological complete response (pcr) (Yerushalmi, Woods, Ravdin et al., Ki-67 in breast cancer: prognostic and predictive potential. Lancet Oncol. (2010): 174-83.), a finding which reflects the fundamental link between tumor replication fraction and activity of cytotoxic agents. 11

Innovation for your breast cancer diagnostics Summary: MammaTyper delivers precise quantitative results for each of the biomarkers: ERBB2 (HER2), ESR1 (ER), PGR (PR) and MKI67 MammaTyper is a highly reproducible test, which avoids disadvantages of intra-/inter-laboratory variation MammaTyper is a precise and reliable test to discriminate the molecular subtypes according to St Gallen Discordant results for MKI67 / Ki-67 were more accurate for RT-qPCR (MammaTyper ) than for IHC according to clinical outcome in the ER positive HER2 negative group MammaTyper is an easy to handle test which allows same day results RT-qPCR is less time-consuming compared to IHC FISH References 150529-01-EN Rev. 1.1 1) Goldhirsch et al., Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol.(2013);24 (9): 2206-2223. 2) Noske et al., Comparison of different approaches for assessment of HER2 expression on protein and mrna level: prediction of chemotherapy response in the neoadjuvant GeparTrio trial. Breast Cancer Res. Treat (2011): 109-17. 3) Laible et al., Technical validation of an RT-qPCR in vitro diagnostic test system for the determination of breast cancer molecular subtypes by quantification of ERBB2, ESR1, PGR and MKI67 mrna levels from formalin-fixed paraffin-embedded breast tumor specimens. submitted. 4) Wirtz et al., Biological subtyping of early breast cancer: A study comparing RT-qPCR with immunohistochemistry. Breast Cancer Res. Treat. (2016): DOI 10.1007/s10549-016-38-7 5) Sinn et al., Comparative analysis of quantitative IHC with automated scoring versus reverse transcription quantitative real-time PCR on FFPE tissue samples for the assessment of ER, PR and Ki-67 labeling index and the prediction of pathological complete response in breast cancer. submitted. MammaTyper and RNXtract are registered trademarks of BioNTech Diagnostics GmbH and STRATIFYER Molecular Pathology GmbH. BioNTech Diagnostics GmbH An der Goldgrube 12 55131 Mainz, Germany Tel: 49 6131 9084 2001 Fax: 49 6131 9084 39 2001 E-Mail: service@sales.biontech.de Visit us online: biontech.de mammatyper.com