1. American Society of Clinical Oncology 2009 Final STEPP results of prophylactic versus reactive skin toxicity (ST) treatment (tx) for panitumumab (pmab)-related ST in patients (pts) with metastatic colorectal cancer (mCRC) Edith P. Mitchell, 1 Mario Lacouture, 2 Heather Shearer, 3 Nicholas Iannotti, 4 Bilal Piperdi, 5 Madhavan Pillai, 6 Feng Xu, 7 Mohamed Yassine 7 1 Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA; 2 Northwestern University, Chicago, IL; 3 Piedmont Hematology Oncology Associates PLLC, Winston-Salem, NC; 4 Hematology Oncology Associates of Treasure Coast, Port Saint Lucie, FL; 5 University of Massachusetts Medical Center, Worcester, MA; 6 Virginia Oncology Care PC, Richlands, VA; 7 Amgen Inc., Thousand Oaks, CA 2 Northwestern University, Chicago, IL; 3 Piedmont Hematology Oncology Associates PLLC, Winston-Salem, NC; 4 Hematology Oncology Associates of Treasure Coast, Port Saint Lucie, FL; 5 University of Massachusetts Medical Center, Worcester, MA; 6 Virginia Oncology Care PC, Richlands, VA; 7 Amgen Inc., Thousand Oaks, CA

2. American Society of Clinical Oncology 2009 Introduction Colorectal cancer is the third most common cancer worldwide, with approximately 1 million new cases diagnosed annually 1 Colorectal cancer is the third most common cancer worldwide, with approximately 1 million new cases diagnosed annually 1 Panitumumab, a fully human monoclonal antibody targeting the epidermal growth factor receptor (EGFR), is approved in the US as monotherapy for mCRC after disease progression on standard chemotherapy and in the EU and Canada as monotherapy in patients with wild-type KRAS tumor status 2,3 Panitumumab, a fully human monoclonal antibody targeting the epidermal growth factor receptor (EGFR), is approved in the US as monotherapy for mCRC after disease progression on standard chemotherapy and in the EU and Canada as monotherapy in patients with wild-type KRAS tumor status 2,3 Skin toxicities are the most common treatment ‑ related side effects of anti ‑ EGFR therapy 3,4 Skin toxicities are the most common treatment ‑ related side effects of anti ‑ EGFR therapy 3,4 The STEPP study is the first prospective study to examine differences between prophylactic and reactive skin treatment for skin toxicities associated with any EGFR inhibitor The STEPP study is the first prospective study to examine differences between prophylactic and reactive skin treatment for skin toxicities associated with any EGFR inhibitor This is the final analysis of STEPP on safety and efficacy of panitumumab by skin treatment and KRAS status This is the final analysis of STEPP on safety and efficacy of panitumumab by skin treatment and KRAS status

3. American Society of Clinical Oncology 2009 Study Objectives To estimate the difference in the incidence of specific ≥ grade 2 skin toxicities of interest between patients receiving prophylactic or reactive skin treatment arms during the 6-week skin treatment period To estimate the difference in the incidence of specific ≥ grade 2 skin toxicities of interest between patients receiving prophylactic or reactive skin treatment arms during the 6-week skin treatment period Secondary objectives: Secondary objectives: – Assessments of other skin toxicity events during the 6-week skin treatment period – Safety and efficacy of panitumumab plus irinotecan- based chemotherapy by KRAS status

4. American Society of Clinical Oncology 2009 Methods: Study Design Phase 2, multicenter, open-label, randomized clinical trial conducted in the US Phase 2, multicenter, open-label, randomized clinical trial conducted in the US Patients received either pmab 6 mg/kg/FOLFIRI Q2W or pmab 9.0 mg/kg/ irinotecan chemotherapy Q3W by investigators choice Patients received either pmab 6 mg/kg/FOLFIRI Q2W or pmab 9.0 mg/kg/ irinotecan chemotherapy Q3W by investigators choice Tumor response was assessed by central review using modified Response Evaluation Criteria in Solid Tumors (RECIST) at weeks 9 and 13, and Q8W thereafter for patients on the Q2W regimen, and at weeks 10, 14, 22, and then Q9W thereafter for patients on the Q3W regimen Tumor response was assessed by central review using modified Response Evaluation Criteria in Solid Tumors (RECIST) at weeks 9 and 13, and Q8W thereafter for patients on the Q2W regimen, and at weeks 10, 14, 22, and then Q9W thereafter for patients on the Q3W regimen Within each chemotherapy stratum, patients were randomized in a 1:1 ratio to either the prophylactic or reactive skin treatment Within each chemotherapy stratum, patients were randomized in a 1:1 ratio to either the prophylactic or reactive skin treatment Patient-reported Quality of Life was assessed using the Dermatology Life Quality Index (DLQI) at screening during weeks 2 to 7, and at the week 13 or 14 visit, depending on treatment schedule Patient-reported Quality of Life was assessed using the Dermatology Life Quality Index (DLQI) at screening during weeks 2 to 7, and at the week 13 or 14 visit, depending on treatment schedule Patients continued on-study until completion of the post-treatment follow-up period, death, withdrawal of full consent, or lost to follow-up Patients continued on-study until completion of the post-treatment follow-up period, death, withdrawal of full consent, or lost to follow-up

5. American Society of Clinical Oncology 2009 STEPP Study Schema Prophylactic Skin Treatment SCREENINGSCREENINGSCREENINGSCREENING RANDOMIZATIONRANDOMIZATIONRANDOMIZATIONRANDOMIZATION PROGRESSIONPROGRESSIONPROGRESSIONPROGRESSION OVERALLOVERALLSURVIVALSURVIVALFOLLOWFOLLOWUPUPOVERALLOVERALLSURVIVALSURVIVALFOLLOWFOLLOWUPUP SAFETYSAFETYFOLLOWFOLLOWUPUPSAFETYSAFETYFOLLOWFOLLOWUPUP SKINSKINTOXICITYTOXICITYEVALUATIONbEVALUATIONbSKINSKINTOXICITYTOXICITYEVALUATIONbEVALUATIONb TUMORTUMORRESPONSERESPONSEEVALUATIONEVALUATIONTUMORTUMORRESPONSERESPONSEEVALUATIONEVALUATION Q8W or Q9W n = 95 n = 48 n = 47 Reactive Skin Treatment 1:1 a a Stratified by chemotherapy regimen (investigator choice): either FOLFIRI and panitumumab Q2W regimen or irinotecan and panitumumab Q3W regimen b Skin treatment period: Weeks 1 to 6 (start day - 1) Skin assessment period: QW from weeks 1 to 7 Skin assessment period: QW from weeks 1 to 7

6. American Society of Clinical Oncology 2009 Skin Treatment Prophylactic skin treatment regimen administered weeks 1 to 6 (beginning day 1): Prophylactic skin treatment regimen administered weeks 1 to 6 (beginning day 1): Skin moisturizer – apply to face, hands, feet, neck, back, and chest daily in the morning upon rising Skin moisturizer – apply to face, hands, feet, neck, back, and chest daily in the morning upon rising Sunscreen (PABA free, SPF ≥ 15, UVA/UVB protection) – apply to exposed skin areas before going outdoors Sunscreen (PABA free, SPF ≥ 15, UVA/UVB protection) – apply to exposed skin areas before going outdoors Topical steroid (1% hydrocortisone cream) – apply to face, hands, feet, neck, back, and chest at bedtime Topical steroid (1% hydrocortisone cream) – apply to face, hands, feet, neck, back, and chest at bedtime Doxycycline 100 mg BID Doxycycline 100 mg BID Per investigator discretion, a reactive skin treatment was administered anytime during weeks 1-6 Per investigator discretion, a reactive skin treatment was administered anytime during weeks 1-6 From week 7 and thereafter, investigators had the option to continue patients on the assigned skin treatment regimen From week 7 and thereafter, investigators had the option to continue patients on the assigned skin treatment regimen

7. American Society of Clinical Oncology 2009 Key Eligibility Criteria ≥ 18 years of age ≥ 18 years of age Unresectable metastatic adenocarcinoma of the colon or rectum that cannot, per investigator, be cured by surgical resection at the time of randomization Unresectable metastatic adenocarcinoma of the colon or rectum that cannot, per investigator, be cured by surgical resection at the time of randomization Measurable disease per modified RECIST Measurable disease per modified RECIST Failure of first-line treatment containing fluoropyrimidine and oxaliplatin-based chemotherapy with or without bevacizumab for mCRC Failure of first-line treatment containing fluoropyrimidine and oxaliplatin-based chemotherapy with or without bevacizumab for mCRC – Treatment failure was defined as progression of disease or toxicity-based from fluoropyrimidine, oxaliplatin, and/or bevacizumab ECOG performance status of 0 or 1 ECOG performance status of 0 or 1 No prior irinotecan use for the treatment of mCRC No prior irinotecan use for the treatment of mCRC

8. American Society of Clinical Oncology 2009 Statistical Analysis This is the final analysis of the STEPP study This is the final analysis of the STEPP study Skin toxicity, anti-tumor efficacy, and safety endpoints were analyzed on the primary analysis set including all randomized patients Skin toxicity, anti-tumor efficacy, and safety endpoints were analyzed on the primary analysis set including all randomized patients Analyses by KRAS were defined as a subset of patients with evaluable KRAS status Analyses by KRAS were defined as a subset of patients with evaluable KRAS status Quality of life data were based on PRO analysis set defined as a subset of patients with baseline and at least one post- baseline PRO results Quality of life data were based on PRO analysis set defined as a subset of patients with baseline and at least one post- baseline PRO results Logistic regression model was employed to estimate the treatment effect on the binary endpoints of interest. The Odds Ratio (OR) and 95% CI using the Wald method were provided Logistic regression model was employed to estimate the treatment effect on the binary endpoints of interest. The Odds Ratio (OR) and 95% CI using the Wald method were provided AEs (adverse events) were grouped by using MedDRA 9.0. The severity of AEs was coded according to the NCI CTCAE v. 3.0 with modification for Dermatology Toxicity Grading for skin related toxicities AEs (adverse events) were grouped by using MedDRA 9.0. The severity of AEs was coded according to the NCI CTCAE v. 3.0 with modification for Dermatology Toxicity Grading for skin related toxicities

9. American Society of Clinical Oncology 2009 Table 1: Patient Disposition Prophylactic Skin Treatment N = 48 Reactive Skin Treatment N = 47 Patients who ended second-line treatment – n (%) 48 (100) 47 (100) Reason for ending treatment – n (%) Disease progression Disease progression 30 (63) 28 (60) Adverse event Adverse event 5 (10) 5 (11) Patient request Patient request 5 (10) 6 (13) Death Death 3 (6) 2 (4) Other Other 5 (10) 6 (13) Median follow-up time 1, weeks 31.040.7 1 Follow-up time is calculated as the randomization date to the last on-study or long-term follow-up visit.

10. American Society of Clinical Oncology 2009 Prophylactic Skin Treatment N = 48 Reactive Skin Treatment N = 47 Sex – n (%) Men 32 (67) 26 (55) Race – n (%) White or Caucasian 34 (71) 40 (85) Black or African American 6 (13) 5 (11) Hispanic or Latino 5 (10) 1 (2) Other 3 (6) 1 (2) Age – years, median (min, max) 60 (24, 84) 61 (40, 86) ECOG performance status – n (%) 0 34 (71) 30 (64) 1 12 (25) 17 (36) 2 2 (4) 0 (0) Primary tumor type – n (%) Colon 34 (71) 28 (60) Rectal 14 (29) 19 (40) Number of metastatic sites – n (%) 1 18 (38) 17 (36) > 1 30 (63) 30 (64) Table 1: Demographics and Disease Characteristics (cont’d)

11. American Society of Clinical Oncology 2009 Primary Endpoint - Incidence of Grade 2 or Higher Skin Toxicities 1 in Prophylactic vs Reactive Skin Treatment Arms (During the Skin Treatment Period, Final Analysis) Prophylactic Skin Treatment N = 48 Reactive Skin Treatment N = 47 Patients with grade 2 or higher skin toxicity – n (%) 2 14 (29) 29 (62) Odds Ratio 3 (95% CL) 0.3 (0.1, 0.6) 0.3 (0.1, 0.6) Grade 2 – n (%) 11 (23) 19 (40) Grade 3 – n (%) 3 (6) 10 (21) 1 Specific skin toxicities of interest per protocol 2 There were no grade 4 skin toxicities during the skin treatment period 3 Odds ratio is estimated from a logistic regression model including treatment (prophylactic vs reactive) that includes an adjustment for chemotherapy stratum (Q2W vs Q3W)

12. American Society of Clinical Oncology 2009 Probability of Grade 2 or Higher Skin Toxicity 1 by Time on the Study 1 Specific skin toxicities of interest per protocol

13. American Society of Clinical Oncology 2009 Best Overall Response 1 and Progression-free Survival By Skin Treatment Group Central Review Prophylactic N = 48 Reactive N = 47 Best overall response – n (%) 7 (15) 5 (11) Complete response 0 (0) Partial response 7 (15) 5 (11) Stable disease 24 (50) 25 (53) Disease control 31 (65) 30 (64) Disease progression 9 (19) 10 (21) Not done or unevaluable 8 (17) 7 (15) Progression free survival – KM Median (95% CI) Months 4.7 (2.9 - 6.0) 4.1 (2.9 - 6.2) 1 By independent central review and confirmed by a follow-up assessment no less than 28 days after the criteria for response were first met

14. American Society of Clinical Oncology 2009 Disposition: KRAS subsets Patients enrolled n (%) Wild-type KRAS N = 49 Mutant KRAS N = 38 Total patients ended treatment – n (%) 49 (100) 38 (100) Reason for ending treatment period – n (%) Disease progression 23 (47) 28 (74) Death 4 (8) 0 (0) Protocol deviation 4 (8) 0 (0) Adverse events 5 (10) 5 (13) Patient request 8 (16) 3 (8) Consent withdrawn 0 (0) 1 (3) Administrative decision 3 (6) 1 (3) Other 2 (4) 0 (0) Follow-up time (weeks) Median39.333.2 Range 0.7, 96.4 1.7, 96.6

15. American Society of Clinical Oncology 2009 Best Overall Response by Central review: KRAS subsets Wild-type KRAS N = 49 Mutant KRAS N = 38 Best overall response – n (%) a Complete response 0 (0) Partial response 8 (16) 3 (8) Stable disease 26 (53) 21 (55) Disease control 34 (69) 24 (63) Disease progression 10 (20) 6 (16) Not done or unevaluable 4 (8) 6 (16) a Responses were confirmed at least 4 weeks after response criteria were first met

16. American Society of Clinical Oncology 2009 Events Median (months) N (%) (95 % CI) KRAS Wild-type34 (69%)5.5 (4.0 - 6.8) KRAS Mutant 26 (68%)3.3 (2.9 - 5.2) HR = 0.8 (95% CI: 0.4 - 1.3) Central Review of Progression Free Survival by KRAS Status:

17. American Society of Clinical Oncology 2009 Best Overall Response 1 and Progression-free Survival by KRAS Status and Skin Treatment Group 2 Central Review Wild-type KRAS Mutant KRAS Prophylactic N = 23 Reactive N = 26 Prophylactic N = 21 Reactive N = 17 Best overall response – n (%) 4 (17) 4 (15) 2 (10) 1 (6) Complete response 0 (0) Partial response 4 (17) 4 (15) 2 (10) 1 (6) Stable disease 13 (57) 13 (50) 11 (52) 10 (59) Disease control 17 (74) 17 (65) 13 (62) 11 (65) Disease progression 3 (13) 7 (27) 5 (24) 1 (6) Not done or unevaluable 3 (13) 2 (8) 3 (14) 5 (29) Progression free survival – KM Median (95% CI) Months 6 (4 - 9) 5 (2 - 8) 3 (3 - 5) 3 (3 - 9) 1 Responses were confirmed via central review and confirmed 28 days after the criteria for response was first met 2 This was an unplanned analysis

18. American Society of Clinical Oncology 2009 Nausea Nausea 32 (67) 3 (6) 0 (0) 26 (55) 4 (9) 0 (0) Vomiting Vomiting 22 (46) 3 (6) 3 (6) 0 (0) 17 (36) 4 (9) 0 (0) Fatigue Fatigue 29 (60) 5 (10) 0 (0) 27 (57) 5 (11) 0 (0) Diarrhea Diarrhea 27 (56) 7 (15) 0 (0) 0 (0) 40 (85) 15 (32) 0 (0) Neutropenia Neutropenia 9 (19) 3 (6) 1 (2) 20 (43) 8 (17) 4 (9) Hypomagnesemia Hypomagnesemia 7 (15) 1 (2) 13 (28) 2 (4) 1 (2) Dehydration Dehydration 6 (13) 3 (6) 3 (6) 0 (0) 16 (34) 8 (17) 0 (0) Adverse Events of Interest 1 Week 1 Through the Safety Follow-up Prophylactic Skin Treatment - N = 48 Reactive Skin Treatment - N = 47 AnyGrade Grade 3 Grade 4 AnyGrade Grade 3 Grade 4 Pts with any event – n (%) 48 (100) 20 (42) 5 (10) 47 (100) 25 (53) 10 (21) Dermatitis acneiform Dermatitis acneiform 37 (77) 2 (4) 0 (0) 0 (0) 40 (85) 10 (21) 0 (0) 0 (0) Pruritus 30 (63) 1 (2) 0 (0) 32 (68) 5 (11) 0 (0) Pustular rash Pustular rash 13 (27) 2 (4) 0 (0) 19 (40) 8 (17) 0 (0) Paronychia 8 (17) 0 (0) 1 (2) 17 (36) 3 (6) 0 (0) 1 There were no grade 5 adverse events of interest

19. American Society of Clinical Oncology 2009 Treatment Exposure Prophylactic Skin Treatment Reactive Skin Treatment Total number of panitumumab doses 1 325333 Total panitumumab doses delayed during the study 1 – n (%) 12 (4) 21 (6) 1 Patients who had changes in panitumumab during the study

20. American Society of Clinical Oncology 2009 Dermatologic Toxicities – Grade 3 0% 5% 10% 15% 20% 25% Dermatitis acneiform Pruritus Pustular rash Paronychia Prophylactic Skin Treatment Reactive Skin Treatment Of these commonly observed dermatologic toxicities, there were no grade 4 or 5 events for either skin treatment group % Patients

21. American Society of Clinical Oncology 2009 Non-Dermatologic Toxicities – Grade 3 and 4 Prophylactic Skin Treatment Reactive Skin Treatment Of these commonly observed non-dermatologic toxicities, there were no grade 5 events for either skin treatment group 0% 5% 10% 15% 20% 25% 30% 35% Nausea Gr 3 % Patients Gr 4 Gr 3 Gr 4 Gr 3 Gr 4 Gr 3 Gr 4 Gr 3 Gr 4 Gr 3 Gr 4 Gr 3 Gr 4 VomitingFatigueDiarrheaNeutropenia HypomagnesemiaDehydration

22. American Society of Clinical Oncology 2009 Prophylactic Skin Treatment N = 46 Reactive Skin Treatment N = 44 Mean (SD) DLQI change from baseline to week 3 – points 1.3 (2.6) 4.2 (5.8) Mean (SD) DLQI change from baseline to week 7 – points 2.0 (2.8) 2.6 (4.4) Mean (SD) DLQI Score Change from Baseline to Week 3 and Week 7 1 1 Based on the PRO Analysis Set defined as patients who had a baseline and at least 1 post-baseline DLQI score Patients in the prophylactic skin treatment group reported improved quality of life, especially during weeks 2 to 3 when the median time to first ≥ grade 2 skin toxicity of interest was reached in the reactive skin treatment group

23. American Society of Clinical Oncology 2009 Conclusions The incidence of grade 2 or higher skin toxicities during the 6-week skin treatment period was reduced by more than 50% in the prophylactic treatment group compared with the reactive treatment group The incidence of grade 2 or higher skin toxicities during the 6-week skin treatment period was reduced by more than 50% in the prophylactic treatment group compared with the reactive treatment group Numerical differences in favor of the wild-type KRAS group were observed for all efficacy endpoints Numerical differences in favor of the wild-type KRAS group were observed for all efficacy endpoints Prophylactic and reactive skin treatment regimen had no difference in efficacy based on skin toxicity regimen Prophylactic and reactive skin treatment regimen had no difference in efficacy based on skin toxicity regimen A phase 3 registrational study is currently ongoing to investigate panitumumab with FOLFIRI by KRAS mutational status in second-line mCRC A phase 3 registrational study is currently ongoing to investigate panitumumab with FOLFIRI by KRAS mutational status in second-line mCRC

24. American Society of Clinical Oncology 2009 References 1.Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin. 2005; 55:74-108. 2.Jakobovits A, et al. Nat Biotech. 2007; 25:1134-1143. 3.Vectibix ® Prescribing Information, Amgen Inc. Thousand Oaks, CA; 2007. 4.Perez-Soler R, Saltz L. J Clin Oncol. 2005; 23:5235-5246.