Amgen Inc. v. F. Hoffmann-LaRoche LTD et al

Filing 600

DECLARATION re #588 Opposition to Motion for Summary Judgment of Infringement of '422 Claim 1, '933 Claim 3, and '698 Claim 6 (by Sven-Michael Cords) by F. Hoffmann-LaRoche LTD, Roche Diagnostics GmbH, Hoffmann LaRoche Inc.. (Huston, Julia)

Amgen Inc. v. F. Hoffmann-LaRoche LTD et al

Doc. 600

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 1 of 11

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS AMGEN, INC., Plaintiff, v. F. HOFFMANN-LA ROCHE, LTD., ROCHE DIAGNOSTICS GMBH, and HOFFMANN-LA ROCHE, INC. Defendants. Civil Action No. 05 CV 12237 WGY

DECLARATION OF DR. SVEN-MICHAEL CORDS IN SUPPORT OF ROCHE'S OPPOSITION TO AMGEN'S MOTION FOR SUMMARY JUDGMENT OF INFRINGEMENT OF '422 CLAIM 1, '933 CLAIM 4, AND '698 CLAIM 6 I, Sven-Michael Cords, hereby declare under penalty of perjury that: 1. I am currently Project Manager and Study Director at Bioassay GmbH in Gerlingen,

Germany. I have worked for Bioassay GmbH in this position since 2006. I am responsible for the supervision and conduction of bioassays for pharmaceutical quality control in accordance with the European Pharmacopoeia and the supervision of bioassay development and business development. The Bioassay GmbH labs are located in Ludwigshafen and in Heidelberg, both of which have animal testing facilities. Currently, ten scientists are employed at Bioassay GmbH laboratories. 2. Before joining Bioassay GmbH, I worked for five years on different animal models

during my doctoral thesis and conducted post-doctoral work in cooperation with animal testing sites at university. I subsequently worked for six years at Biomet Merck in Darmstadt, Germany, as the research and development manager in the development of medical devices such as bone substitutes and

1

Dockets.Justia.com

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 2 of 11

coated implant materials. In addition, I was responsible for production of Patient Matched Implants in accordance with GMP at Biomet-Merck in Darmstadt, Germany. 3. I received my MBA in 2006 from the Merck German Partnership program at Pforzheim

University of Applied Science. My MBA Diploma Thesis was in Evaluation of Outsourcing Attractiveness Factors in Biopharmaceutical Manufacturing with Focus on Analytical Services. I also studied veterinary medicine from 1987-1993 at the Veterinary University of Hannover. In July 1993, I passed the third part of the veterinary exam and approbation. (See C.V. attached to this report as Exhibit 1). A list of my publications and presentations is attached to this report as Exhibit 2. 4. Since joining Bioassay GmbH in 2006, I have run bioassay tests as part of my daily

activities at the laboratory in Ludwigshafen. I have tested approximately 1000 samples, measuring the biological activity of Erythropoietin-samples in the parallel line assay according to the European Pharmacopoeia. 5. information. 6. I have been asked by counsel for Defendants1 to describe certain tests I performed at I make this declaration based upon my own personal knowledge and company

Defendants' request relevant to this Court's determination of whether material differences exist between Roche's unique chemical compound, Mircera, and human erythropoietin as claimed by the six patentsin-suit.2 These tests were performed under contract in Ludwigshafen, Germany, at the labs of Bioassay

1

F. Hoffmann-La Roche, Ltd., Roche Diagnostics GmbH, and Hoffmann-La Roche, Inc. (collectively "Roche").
2

U.S. Patent Nos. 5,441,868 (the " `868 patent"), 5,547,933 (the " `933 patent"), 5,618,698 (the " `698 patent"), 5,621,080 (the " `080 patent"), 5,756,349 (the " `349 patent"), and 5,955,422 (the " `422 patent") (collectively "the Amgen patents").

2

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 3 of 11

Labor für biologische Analytik GmbH ("Bioassay GmbH"). All tests were performed under my direct supervision. 7. In the paragraphs below, I set forth a detailed description of the experiments performed

and state that the analyses presented herein faithfully reflect experimental data observed, collected, or measured. 8. I have conducted an independent bioassay experiment with various Mircera and EPO

samples and controls, as instructed by Defendants, using samples prepared and provided by Defendants. In performing the experiments, I have relied on my years of education, research, and experience using animal bioassays as model assays for human systems, and my wealth of experience performing bioassays on EPO samples as part of my regular duties at Bioassay GmbH according to contracts with other clients. 9. The protocol used in these experiments, a modified protocol of the Normomouse

Bioassay, is a reliable system through which to measure the kinetic responses of Roche's erythropoietic stimulating agent, Mircera, on reticulocyte production in mice, just as it is a reliable system to measure the kinetic response of Epoetin beta on reticulocyte production in mice, and to present the differences in stimulatory effect observed with Mircera versus Epoetin beta with the samples tested. I have not been asked to provide an opinion as to the pharmacological and/or pharmacokinetic or pharmacodynamic differences between Mircera and recombinant human erythropoietin according to the claims of the Amgen patents. 10. Based on the results obtained, as shown in Figures 1, 2, and 3 below, it is my opinion

that the bioassay used to generate the data in these figures is a reliable system through which to observe and measure the kinetic activity, namely, reticulocyte production over time, of Roche's Mircera product,

3

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 4 of 11

and that such results with Mircera can be compared to results obtained using the same bioassay performed with Epoetin beta (which I may refer to herein simply as EPO). 11. This conclusion is based upon an understanding of bioassay techniques studying

erythropoietic stimulating agents (ESAs), and my reasoning for reaching this conclusion is set forth below. I. BACKGROUND INFORMATION A. 12. Mircera

I understand that the Mircera drug substance is chemically synthesized by combining a

reactive succinimidyl ester of methoxy PEG-succinimidyl butanoic acid with Epoetin beta at one or more free primary amino groups. I also understand that this reaction results in a molecule with a molecular weight of approximately 60 kDa. As discussed below, I conducted a bioassay demonstrating the kinetic effects in mice of the Mircera drug substance and Epoetin Beta, as indicated by production of reticulocytes over time. B. 13. Bioassays

I have specifically worked with bioassays involving biologically active erythropoietin

using a bioassay that quantitatively counts reticulocyte production in the blood of mice over time, after treatment with samples of EPO. I am highly familiar with analytical methods for quality control in conducting bioassays, and Bioassay GmbH does not conduct such tests exclusively for Roche In fact, our company typically performs over 1000 such assays per year for our clients. Therefore, I am particularly suited for performing the modified version of the Normomouse Bioassay requested by the Defendants, the results of which are described in this report.

4

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 5 of 11

II. 14.

IN VIVO COMPARISON OF MIRCERA TO EPO I conducted certain tests at the direction of Defendants using samples prepared and sent

by Defendants, using a protocol designed by and received from Defendants. In particular, I conducted a bioassay with Mircera, deglycosylated Mircera, EPO and deglycosylated EPO samples, and appropriate controls. The experiment was designed and performed to evaluate the efficacy and reliability of using this particular bioassay for comparing the kinetic differences between the in vivo activity of Mircera and Epoetin beta, based on an increase in reticulocytes in the blood of mice over time, after treatment with the various samples and controls, and to generate useful data regarding the samples by which they can be compared. I provide below an overview of the protocol used, and a presentation of my observations and interpretation of the data collected.

A. 15.

Sample Preparation

The following seven samples were received from Defendants on March 26, 2007 for

study in the bioassay experiment (with sample concentrations in parenthesis): · · · · · · · 16. Epoetin beta (8800 g/mL) Epoetin beta mock-treated (240 g/mL) Epoetin beta deglycosylated (100 g/mL) Mircera (5900 g/mL) Mircera mock-treated (130 g/mL) Mircera deglycosylated (120 g/mL) Control (0 g/mL)

It is my understanding that the samples received and listed above were prepared by

Markus Dembowski, of Roche Pharmaceuticals, Penzberg, Germany (see the Declaration of Markus Dembowski, attached to this report as Exhibit 4). I was not informed of how the samples were prepared, or what, exactly, the samples were, other than what is apparent from the brief descriptions and concentrations listed above.

5

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 6 of 11

17.

After receiving the samples, all samples were diluted to a concentration of 200 ng/mL

each in 1-3 dilution steps, depending on the starting concentration of the particular sample.

B. 18.

Test Procedure Protocol

In order to measure the in vivo responses to the various sample treatments in normal,

healthy male B6D2F1 mice, 7 weeks old, approximately 20g, reticulocyte counts per 30,000 cells were measured at 48, 72, 96, and 120 hours after treatment. 19. A total of 140 mice were used to conduct the bioassay. Twenty mice were treated and

tested for each of the samples. Each group of twenty mice was divided into four subgroups of five mice per day over the course of four days. 20. The mice in the first six groups were each injected subcutaneously with 0.5 mL of the

final solution, containing 100 ng of the respective samples: (1) EPO, (2) EPO K (mock-treated), (3) deglycosylated EPO, (4) Mircera, (5) Mircera K (mock-treated), and (6) deglycosylated Mircera; the seventh group, the control group, was injected subcutaneously with 0.5 mL of the vehicle solution only, which contained no test sample. 21. For each of these seven test groups, five different mice were bled on each of the four

indicated days. Blood samples of 5 L were drawn by puncture of the tail vein of each mouse. 22. The blood was then diluted so that 1 L of blood was present in 1 mL of a 0.15-mM

acridine orange staining solution. The staining time was 3-10 minutes. Reticulocyte count per 30,000 cells was determined for each mouse in the study by FACS-analysis at hours 48, 72, 96, and 120 in accordance with standard procedures using a Becton Dickinson FACScan flow cytometer with loader.

C.

Results

6

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 7 of 11

23.

Figure 1 below represents in table form the results generated by the bioassay protocol

described above, including calculation of the mean, standard deviation, and relative standard deviation for each group. Epoetin Mircera Control beta Start mock DeglycoStart mock DeglycoMouse vehicle material treated sylated material treated sylated 1066 834 498 963 1242 761 238 1 value 937 895 465 924 800 786 344 2 value 1084 960 388 1105 954 928 232 3 value 1116 911 470 1126 1084 951 252 4 value 1263 945 504 993 1045 712 413 5 value mean 1093 909 465 1022 1025 828 296 SD 117 49 46 89 163 106 80 RSD 11 5 10 9 16 13 27 1386 1430 555 2136 2048 1466 401 1 value 1403 1189 641 2006 1921 1424 304 2 value 1383 1460 341 1851 1730 1456 252 3 value 1481 1209 346 2012 1760 1350 428 4 value 1310 1225 524 2083 1800 1311 384 5 value mean 1393 1303 481 2018 1852 1401 354 SD 61 131 133 107 132 68 73 RSD 4 10 28 5 7 5 21 1285 797 459 3526 2149 839 341 1 value 1258 745 333 2783 3015 1489 388 2 value 1350 803 3022 2360 1527 367 3 value 1697 1333 988 594 2957 2359 1170 325 4 value 1070 949 344 2681 2406 1378 259 5 value mean 1259 856 433 2994 2458 1281 336 SD 112 106 122 327 327 283 49 RSD 9 12 28 11 13 22 15 482 392 326 3134 2754 667 401 1 value 412 379 360 2644 3038 652 316 2 value 343 381 437 3801 3168 475 794 3 value 381 358 453 3125 2661 584 337 4 value 440 404 282 3176 2279 570 349 5 value mean 412 383 372 3176 2780 590 439 SD 53 17 73 412 347 77 201 RSD 13 4 20 13 12 13 46 "value" = count of reticulocytes per 30,000 cells Data 96 h: Value 3 EPO deglycosylated is found technical outlier (mouse) and eliminated Figure 1. 7

Time [hrs] 48

72

96

120

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 8 of 11

24. 25.

One technical outlier was eliminated (Epoetin beta Deglycosylated, 3 value, 96 hours). As Figure 1 demonstrates, the relative standard deviation of the reticulocyte counts for

both Epoetin beta and Mircera throughout the study were very low and generally under 15% in those groups which showed a distinct value above the untreated control. 26. Figure 2 below shows a graph corresponding to the reticulocyte counts for each of the

seven groups of mice measured at hours 48, 72, 96, and 120.
Kinetic Study
4000

3500

# Reticulocytes (per 30K cells)

3000

2500

2000

1500

EPO EPO K EPO Deglyc Mircera Mircera K Mircera Degly Control

1000

500

0 48 72 96 120

Time (hrs)

Figure 2. 27. As shown in Figure 2, mice treated with each variant of Mircera showed higher

reticulocyte responses at 72, 96 and 120 h than those treated with the respective EPO variants. Furthermore, Mircera-treated and Mircera mock-treated mice showed a sustained increasing response through the first 120 h, whereas their respective EPO counterparts reached a maximum reticulocyte count at 72 h and thereafter showed decreased reticulocyte counts. Also, deglycosylated Mircera-treated 8

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 9 of 11

mice showed nearly identical reticulocyte counts at 72 h and 96 h compared to the number of reticulocytes observed for EPO-treated mice at 72 h and 96 h, and only slight differences in reticulocyte counts were observed between deglycosylated Mircera and EPO at 48 h and 120 h. Further, after 72 h deglycosylated EPO-treated mice showed similar reticulocyte counts as those from the control group (vehicle alone), i.e. no significantly detectable stimulation of reticulocyte production was observed.
Kinetic Study (Normalized to Control)
3500

3000

# Reticulocytes (per 30K cells)

2500

2000

1500

1000

EPO EPO K EPO Deglyc Mircera Mircera K Mircera Degly

500

0 48 -500 72 96 120

Time (hrs)

Figure 3. 28. Figure 3 above shows the data from Figure 2 normalized to the reticulocyte counts

measured for the control (vehicle alone) mouse group. This graph emphasizes the difference in the response observed in EPO treated mice compared to that seen in Mircera-treated mice. Again, Mircera induced much stronger and prolonged erythropoietic responses, as indicated by reticulocyte counts/30,000 cells, over the 120-hour measurement span.

9

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 10 of 11

29.

The above results indicate that this bioassay is a very reliable and accurate means by

which to evaluate the in vivo kinetic responses induced by treatment with EPO or Mircera, as indicated by increase reticulocyte counts per 30,000 cells over time, and that comparison of the results obtained in this bioassay for these two molecules is possible. 30. The results of this procedure lead me to conclude that this bioassay is thus a reliable

system through which to observe and compare the erythropoietic responses in mice to treatment with EPO or Mircera. 31. I hereby certify that the experimental results described herein faithfully represent my I furthermore declare that all information so

direct observations of each of the tests performed.

generated by me has been disclosed to both parties to the litigation in its entirety. 32. In light of the foregoing analysis, it is my expert opinion that the bioassay I conducted,

which is a modified version of the Normomouse Bioassay, is a reliable and accurate assay to measure the in vivo kinetic responses in mice to EPO or Mircera, and that this assay may also be used to compare reticulocyte kinetic response to these compounds in mice. 33. The results of the experiments described in this declaration show that deglycosylated

Micera (containing the active ingredient CERA) still retains a significant amount of its erythropoietic activity as compared to deglycosylated EPO. Given the significant erythropoietic activity demonstrated by N-deglycosylated Micera, these data strongly indicate that CERA is not an obligate glycoprotein. I declare under penalty of perjury that the foregoing is true and correct. Executed on: June 29, 2007 /s/ Sven Michael Cords Dr. med. vet. Sven-Michael Cords, MBA

10

Case 1:05-cv-12237-WGY

Document 600

Filed 06/29/2007

Page 11 of 11

CERTIFICATE OF SERVICE I hereby certify that this document filed through the ECF system will be sent electronically to the registered participants as identified on the Notice of Electronic Filing (NEF) and paper copies will be sent to those indicated as non registered participants on the above referenced date.

/s/ Keith E. Toms______________

03099/00501 695270.1

11