Osi Pharmaceuticals 8K Regulation FD Disclosure | OSIP 29 Nov 07

Exhibit 99.1

Shaping Medicine, Changing Lives Welcome to OSI's Analyst Day November 29, 2007 Exhibit 99.1

OSI Pharmaceuticals Forward-looking Statements OSI's presentations today contain forward-looking statements. These statements are subject to known and unknown risks and uncertainties that may cause actual future experience and results to differ materially from the statements made. Factors that might cause such a difference include, among others, the FDA review process and other governmental regulation, OSI's and its collaborators' abilities to successfully develop and commercialize drug candidates, competition from other pharmaceutical companies, the ability to effectively market products, and other factors described in OSI Pharmaceuticals' filings with the Securities and Exchange Commission. OSI's presentations also contain select financial guidance for the fiscal year ended December 31, 2007 which OSI provided on October 25, 2007. Please note that OSI is not updating its guidance for 2007 at this time.

Shaping Medicine, Changing Lives Colin Goddard, Ph.D. Chief Executive Officer Analyst Day November 29, 2007 CEO Gold Standard Founding Member Strategic Overview

OSI Analyst Day Agenda Time Activity Speaker 12:30 Welcome Kathy Galante 12:35 Strategic Overview Colin Goddard, Ph.D. 12:50 Oncology Gabriel Leung David Epstein, Ph.D. Karsten Witt, M.D. 2:00 Diabetes & Obesity Anker Lundemose, M.D., Ph.D. Jim McCormack, Ph.D., D.Sc Jonathan Rachman, M.D., Ph.D. 2:55 Financial Commentary Mike Atieh 3:15 Wrap-up / Q&A Colin Goddard, Ph.D. 3:30 Poster Session and Cocktail Reception

OSIP Exiting 2007: Re-Establishing Growth Weathering the Classic "First Product" Transition Years to/from Launch: -1 0 +1 +2 Science Driven Expectation Financial Performance 'Chapter Two' Can Only Really Begin When Financial Performance Catches Up to Science Driven Expectations Short Squeeze Around Br.21 Accentuated Expectations Eyetech Acquisition - Designed to Bridge Revenues And Plug Expectation/Financial performance Gap - Failed Badly and Ended up Accentuating the Gap And Delaying the Recovery Typical Stock Performance Through Transition to an Established Company Creates Challenges Intense Competition, Lack of Infrastructure & Risk Assessment on execution Led Us to Partner Tarceva & Lose Control. Mitigated Somewhat by Outstanding Deal Economics With the Failure of the Eyetech Acquisition We Committed to a Turnaround/Growth Strategy Designed to: Maintain Focus on Tarceva(r) Execution Appropriately Balance Financial Performance With Re- Investment in the Oncology & Diabetes Asset Base by: Re-Aligning Cost-Base and Infrastructure With Revenues (Headcount Reduction, Facilities Rationalization) Highly Disciplined R&D Investments Focused on an Internal Pipeline of Innovative and Differentiated Core Assets and Technology Platforms Divesting and Monetizing Non-Core Assets (GKA Deal With Lilly, PDGF Aptamer Out-License, Eyetech Divestiture Planned)

Delivering Financial Performance Turning the Corner to Profitability in 2007 Tarceva Sales $MM (except per share amounts) 3Q2007 Nine Months 2007 Nine Months 2006 Guidance U.S. Sales $101.00* $304.00 $296.00 ROW Sales $125.00 $331.00 $164.00 Global Sales $226.00 $635.00 $460.00 >$860 OSI P&L Highlights - Continuing Operations OSI P&L Highlights - Continuing Operations $MM (excluding EPS) 3Q2007 Nine Months 2007 Nine Months 2006 Guidance Revenues $100.37 $256.72 $171.71 >$335 Research & Development $ 29.49 $ 87.38 $ 86.25 $130.00 SG&A $ 24.82 $ 73.98 $ 78.86 $100.00 Total Expenses $ 64.41 $176.34 $171.93 Net Income - Continuing Ops $35.96 $ 84.87 ($1.90) Dil. EPS - Continuing Ops $ 0.59 $ 1.41 ($0.03) $1.65 Dil. EPS - Overall $ 0.49 $ 0.95 ($6.31) Cash and Cashflow Cashflow from Continuing Operations $95.00 Period Ending Cash $286.00 * Includes $13 million in reserve adjustments

OSIP: What to Expect in 2008/9 Laying the Foundation for Sustained Growth "Continued Highly Disciplined R&D Investment Allowing Us to Further Leverage Tarceva Revenues and DP-IV Royalty Top-line in 2008 While Delivering Second Half SATURN & BETA Data and Continuing to Nurture a High Quality Early Pipeline in Oncology and Diabetes/Obesity." We Anticipate That the Core Business Should Generate Approximately $2 per Share in 2008 Independent of Any Supplemental Out-Licensing or Partnering Revenues

OSIP: What to Expect in 2008-9 Next 11/2 - 2 Years Can Deliver Substantial Upside Convince the world of the robust continued global growth of Tarceva Tarceva as a "$2bn+" product & not a "peak $1bn product with no growth" SATURN, BETA and ATLAS can get us there and they read-out over the next 12-18 months Deliver sustained financial performance & strong EPS growth Focused & disciplined R&D spend and effectively monetizing non-core assets are key to maintaining an effective balance between re-investment and financial performance Establish a follow-on portfolio with differentiated core assets under our control that provide an appreciable step-up in value High-potential early programs OSI-906 & PSN821 have clear 'step-up' potential OSI-027 & PSN602 will emerge into clinical arena Disciplined choices for 2008 OSI-930, OSI-632 will either be differentiated from crowded VEGFR field or partnered Validate the excellence of our core science EMT/Diabetes engine We Have A High Probability of Delivering a Substantial Step-Up in the Value and Perception of this Company if We:

Disciplined R&D Spend: Responding to Data A Change in Status for Our Internal DP-IV Program DP-IV inhibitors have two potential liabilities Chronic co-substrate inhibition (Substance P, PYY) DP8/9 selectivity (FDA mandated 3-month primate toxicity study) PSN9301 designed to manage both via short acting half-life and prandial dosing Limited DP8/9 selectivity off-set by slow cellular uptake & low systemic exposure Promising Phase IIa data and apparent window in 28-day primate tox BUT emerging skin lesions in 3-month tox remove differentiation advantage Disappointing Data From Ongoing 3-Month Primate Tox Study Undermines Our Differentiation Thesis and Reason for Continued Investment but Reduces Anticipated 2008/9 R&D Spend by Approximately $40MM Active GLP-1 Levels Promising Phase IIa Data Supported Prandial Dosing Thesis PSN9301 was well tolerated and demonstrated promising efficacy

DP-IV Class: Still a Major Growth Area DP-IV Patent Estate & License Portfolio Ownership of patent estate for the use of DP-IV inhibitors for the treatment of diabetes, as monotherapy and in combination is a major secondary revenue stream for OSI Acquired as part of $35MM acquisition from Probiodrug in 2004 12 large pharma companies have signed license agreements including Merck, Novartis, BMS (Astra-Zeneca) and Roche Januvia(tm) launched in Fall 2006, Janumet (tm) approved in March 2007, Galvus(r) approved in EU, Saxagliptin and alogliptin currently in Phase III trials Januvia has excellent DP8/9 selectivity We believe it will be the market leader & drive OSI royalty revenues Annual Royalty Stream From DP-IV Patent Estate Would Approximate $100 Million for each $5 Billion of Sales for DP-IV Market

OSIP Portfolio in 4Q2007 Tarceva Franchise & a Quality Emerging Pipeline PRODUCT Pre-clinical/ IND TRACK PHASE I PHASE II PHASE III Marketed Tarceva: 2nd/3rd Line NSCLC Pancreatic Cancer 1st Line NSCLC Adjuvant NSCLC 1st Line Ovarian Cancer Other Indications Oncology: OSI-906 (IGF-1R) OSI-027 TORC1/TORC2 (mTOR) OSI-930 (c-kit/VEGFR-2) OSI-632 (VEGFR) Diabetes/Obesity: PSN602 (S1RUP) PSN821 (GPR119 Agonist) PSN010 (GKA) - Lilly

Tarceva: Solid Performance Since Launch Robust Global Growth of the Brand Continues Quarterly Tarceva Net Sales First 3 quarters of 2007 Worldwide Tarceva Net Sales Were $636 million OSI Co-promotes with Genentech in US (equal profit share economics) Roche Commercializes ex-US (OSI Royalty) Guidance as of October 25, 2007 2007 Worldwide Tarceva Sales Should exceed $860 million (32% increase over 2006)

Tarceva: Competitively Positioned Major Potential "Step-Ups" in Value On the Horizon Major Near Term Phase III Data Read-outs: Longer Term Phase II/III Data Read-outs: Study Indication/Use Data SATURN NSCLC: Tarceva Monotherapy as Maintenance Following Front-line Chemotherapy, 850 pts - Label Enabling 2H2008 BETA-LUNG NSCLC: Tarceva/Avastin in Combination as 2nd-line Therapy (non-squamous), 650 pts - Label Enabling 2H2008 ATLAS NSCLC: Tarceva/Avastin as Maintenance Following Front-Line Avastin/Chemotherapy, 1,150 pts 2009 Study Indication/Use Data Never-Smokers NSCLC: Tarceva in Combination with Chemotherapy vs. Tarceva as Monotherapy in Front-line Never Smokers (CALGB), 180 pts 2009 OVARIAN Ovarian: Tarceva Monotherapy as Maintenance Following Front-line Chemotherapy, 830 pts - (EORTC) 2010 RADIANT Stage I-IIIa NSCLC: Tarceva as Maintenance Following Adjuvant Surgery/Chemotherapy, 1,000 pts - Label Enabling 2014/5

Tarceva: Competitively Positioned Competitive Landscape Holds Substantially Less Risk NSCLC studies missing endpoints or halted in last 18 months Velcade Telcyta Iressa (Stage III NSCLC Maintenance) Erbitux in front-line combination with Carbo-Taxol Only Erbitux with less common Navelbine/Cisplatin combination was positive in front-line NSCLC Data yet to be presented Pancreatic cancer studies missing endpoints or halted in last 18 months Avastin + Gemcitabine Erbitux + Gemcitabine Xeloda + Gemcitabine Oxaliplatin + Gemcitabine Eight Out of Nine Major Clinical Trials of Competitive Agents Failed Over the Last Two Years, Which - While Disappointing for Cancer Patients - Serves to Demonstrate Tarceva's Unique Value

OSI Oncology R&D Exiting 2007 EMT Platform: Leadership in Emerging Field Molecular Targeted Therapies Guided by EMT Biology PTK/EMT inhibitor program has advanced OSI-906 and OSI-027 Historically 5 MTTs advanced to clinic from PFE collaboration (including Tarceva) EMT interest stimulated by sub-set analysis of failed TRIBUTE data suggesting E-Cadherin + (epithelial like) tumors benefited greatly from Tarceva + Chemo regimen despite failed overall trial Oncology Research Programs Focus on Epithelial-to-Mesenchymal Transition (EMT) to Exploit Key Molecular Targets Future synergistic Molecular Targeted Therapy combinations with Tarceva Building platform of IP, know-how & tools AVEO alliance could realize novel targets, biomarkers & EMT-like animal models EMT Heterogeneity Requires Combination Therapy

OSI Oncology R&D Exiting 2007 EMT Platform: Leadership in Emerging Field Molecular Targeted Therapies Guided by EMT Biology PTK/EMT Inhibitor Program Has Advanced OSI-906 and OSI-027 Historically 5 MTT's advanced to clinic from PFE collaboration (including Tarceva) EMT interest stimulated by sub-set analysis of failed TRIBUTE data suggesting E-Cadherin + (epithelial like) tumors benefited greatly from Tarceva + Chemo regimen despite failed overall trial Oncology Research Programs Focus on Epithelial-to-Mesenchymal Transition (EMT) to Exploit Key Molecular Targets Future Synergistic Molecular Targeted Therapy Combinations With Tarceva Building Platform of IP, know-how & tools AVEO alliance could realize novel targets, biomarkers & EMT like animal models EMT Heterogeneity Requires Combination Therapy Our Oncology Thesis: "Delivering an Optimized Drug for a Molecular Target to Selected Patients in the Best Biologically Driven Combination With Other Molecular Targeted Therapies Will be Essential to Delivering the Substantial Efficacy Improvements That Will be Necessary to Command Premium Pricing and Drive Market Share in the Increasingly Crowded and Price Sensitive Oncology Arena of the Future."

OSI Development Portfolio in 4Q2007 Quality Candidates in Oncology & Diabetes/Obesity Molecule Disease Target & Differentiation "Step-Up" in Value Data OSI-906 Oncology Potential "First-in-Class" small molecule IGF-1R inhibitor in Phase I trials with demonstrated pre-clinical synergy with Tarceva & potential utility in NSCLC, CRC, prostate & breast cancer Manageable hyperglycemia in Phase I & efficacy in Phase II 2H2009 PSN821 Diabetes/Obesity Potential "Best-in-Class" GPR119 agonist on IND-track. Increases GLP-1 & beta cell cAMP leading to glucose control, delayed gastric emptying & appetite suppression Safety & initial efficacy in Phase I/IIa studies 2H2009 OSI-027 Oncology Potential "First-in-Class" small molecule TORC1/TORC2 inhibitor within EMT thesis - Expected to enter clinic in 1Q2008 Phase I program synergy with Tarceva & potential fast-track in ovarian/4EBP1 1H2009 PSN602 Obesity Potential "Best-in-Class" Serotonin/noradrenaline reuptake inhibitor w/5HT1a agonism - differentiates vs Meridia (low CV side-effects) Phase I confirms lack of acute CV effects & Phase II shows superior weight loss 2H2008 & 2009

OSI Development Portfolio in 4Q2007 Programs With Differentiation Challenges Molecule Disease Target & Profile Challenges Data PSN9301 Diabetes Prandially dosed, short-acting DP-IV inhibitor w/activity in Phase IIa trials & potential for inter-prandial sparing weight-loss & side-effect benefits via sparing other DP-IV substrates Insufficient in-vivo selectivity versus DP8/9 (Final Monkey Tox ongoing) 1Q2008 OSI-930 Oncology Small molecule VEGFR/c-kit inhibitor with early indications of activity in Phase I trials Crowded VEGF/VEGFR field 1H2008 OSI-632 Oncology Small molecule VEGFR/FGFR inhibitor returned from Pfizer & being evaluated in EMT models Crowded VEGF/VEGFR field 1H2008 Programs Will Be Run Through Key Data Points & Then Partnering/Out-licensing or Continuation Decision Made

Partnering/Divestiture of Non-Core Assets A Source of Valuable Supplemental Revenue 2007 Licensing of PSN010 (small molecule GKA) to Lilly Core asset partnered due to financial constraints $25MM up-front, up to $360MM milestones & tiered double-digit royalties Licensing of PDGF-aptamer to Ophthotech $7MM cash & stock plus milestones & royalties Monetization of prior TGF-beta 3 anti-scarring license to Renovo Share of economics from Renovo deal with Shire contributed $6MM net in 3Q2007 2008 Divestiture of remaining eye business assets (Macugen) Divestiture of historical asthma/COPD assets Spin-out envisaged for portfolio of IND-track assets and GPCR platform Potential partnering of OSI-930, OSI-632 & PSN9301 Should emerging data not support necessary differentiation to justify development as a core asset & data support effective partnering Potential partnering of non-obesity uses of selected S1RUP program agents

OSIP Exiting 2007 Summary Strongly profitable Company with anticipated 2007 revenues of >$335MM and earnings of >$1.65 from continued operations Core business may generate ~$2 per share in 2008 Flagship asset in Tarceva with strong global growth and significant market expansion opportunities on the horizon Label enabling SATURN & BETA lung trials read-out in 2H2008 Emerging pipeline of core development assets with clear paths to create "step-ups" in value PSN821 & PSN602 in Diabetes/Obesity OSI-906 & OSI-027 in Oncology Corporate strategy that seeks to build shareholder value by appropriately balancing financial performance with re-investment in the core Oncology & Diabetes/Obesity asset base via: Re-aligning cost-base and infrastructure with revenues Highly disciplined R&D investments focused on an internal pipeline of innovative and differentiated core assets and technology platforms Divesting and monetizing non-core assets

Gabriel Leung President, (OSI) Oncology Executing the Tarceva Program

Tarceva Potential Growth Drivers US Growth Drivers Submission of smoking MTD data for NDA Refined efficacy messaging (e.g. rash) Especially in pancreatic cancer Sales force redeployment SATURN & BETA-Lung data in 2H08 Ex-US Growth Drivers Japan launch Pancreatic launch in EU UK revisited SATURN & BETA-Lung data in 2H08 Long-Term Growth Drivers MTT Cocktails: Avastin + T, Sutent + T, Nexavar + T, etc. Other tumor types: Ovarian Cancer, CRC, HCC Adjuvant NSCLC (RADIANT) Treatment beyond progression

Tarceva Smoker MTD Data Summary MTD in NSCLC patients who continue to smoke (300 mg/d) is higher than the MTD in non-smoking NSCLC patients (150 mg/d) Significantly higher Tarceva exposure at 300 mg/d vs 150 mg/d in current smokers Cmin in current smokers at 300 mg/d is similar to historical data (BR.21) for never/former smokers at 150 mg/d Preliminary data in this small group of patients suggests the 300 mg/d Tarceva dose is tolerable in current smokers

Tarceva and Biomarkers Survival by Rash Grade: BR.21 and PA.3 Rash Evaluable Population (>28 day survival) Survival Probability Survival (Months) 0 6 12 18 24 30 36 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.0 Median: 3.42 Months 95% CI: (2.53, 4.14) Median: 7.18 Months 95% CI: (6.08, 9.40) Median: 11.07 Months 95% CI: (9.95, 12.58) Rash Grade HR 95% CI P value 2+ vs 0 0.47 (0.34, 0.64) <0.001 1 vs 0 0.90 (0.66, 1.23) 0.507 2+ vs 1 0.52 (0.38 0.71) <0.001 BR.21 - Tarceva Arm PA.3 - Tarceva Arm Grade 0 (N=89) Grade 1 (N=136) Grade 2+ (N=219) HR: 0.56* 95% CI: (0.49, 0.64) p = <0.001 Survival (Months) Grade 0 (N=74) Median: 5.4 months Grade 1 (N=92) Median: 5.7 months Grade 2+ (N=88) Median: 10.8 months Survival Probability

(OSI) Oncology Sales Infrastructure An Enhanced Role in Tarceva Promotion In Mid-2007 OSI and Genentech Agreed on a Dedicated Tarceva Sales Force OSI Covers 87% of Tarceva Prescribing Universe and 84% of Zip Codes and Supplies 50% of Total U.S. Reps Sales Specialists include numerous award winners from Novartis, AstraZeneca, Aventis, Amgen, Sanofi, Berlex, Immunex, Eli Lilly, Merck, Ortho, TAP and other leading pharma, biotech and oncology companies. 5 Regions, 50 Sales Specialists

SATURN & BETA The Next Step Up for Tarceva If both studies are successful... Overcoming the practice economic hurdle SATURN: maintenance Tarceva not displacing chemo use BETA: the added Avastin is a reimbursed Part B drug Overcoming the "superior efficacy" hurdle SATURN: maintenance Tarceva > waiting for progression BETA: AT > T (> chemo alone)

SATURN Maintenance 1st-line NSCLC Eligible pts: 1st line Stage IIIB / IV NSCLC Placebo Tarceva Collect tissue Treatment to PD, unacceptable toxicity or death Primary Endpoint: Progression-free survival (Overall population and EGFR+ patients by IHC) N=850 Stratify for EGFR IHC +/- Smoking Status 1st-line plat.-based chemo x 4 cycles CR/PR/SD pts PD patients enter the second line study

SATURN vs BR.21 Moving Tarceva Upstream 1st-line Chemo PD CR/PR/SD Wait for PD BR.21 1st-line Chemo PD CR/PR/SD SATURN TITAN IHC+

BR.21Subset Data Patients with 1 Prior Regimen and CR/PR/SD HR = 0.62 (95% CI, 0.47 - 0.82) Tarceva median PFS = 3.45 mon (n=184) Placebo median PFS = 1.84 mon (n=85)

BR.21Subset Data PFS According to EGFR Protein Expression EGFR IHC+ EGFR IHC- HR = 0.50 (95% CI, 0.36 - 0.69) Tarceva (n=117) Placebo (n=67) HR = 0.80 (95% CI, 0.55 - 1.16) Tarceva (n=93) Placebo (n=48) Tarceva median PFS = 3.68 mon Placebo median PFS = 1.81 mon Tarceva median PFS = 1.87 mon Placebo median PFS = 1.82 mon Interaction P-value = 0.15

Phase III N = 650 Recurrent NSCLC Rx'd brain mets and peripheral squamous eligible 1 prior regimen ECOG PS 0 - 2 Consent for archival Dx tissue required Tarceva + placebo Tarceva + Avastin 1:1 Randomization Stratified by PS, Smoking Hx, Gender, Study sites Primary Endpoint: Overall Survival BETA-Lung AT vs T in 2nd-Line NSCLC

Tarceva/Avastin The Leading All Targeted Therapy Combination Regimen Protect 2nd-line NSCLC business Could displace chemotherapy in 1st-line NSCLC Potential in other tumor types Attractive for chronic therapy, but can be further improved with oral anti-VEGF agents A/T A/Chemo Chemo N 39 40 41 PFS 4.4 mos 4.8 mos 3.0 mos HR (PFS) 0.72 0.66 1 6 mos PFS 33.6% 30.5% 21.5% Med Surv 13.7 mos 12.6 mos 8.6 mos 1-Yr Surv 57.4% 53.8% 33.1% SAEs 33% 41% 55% Drug D/C 5 11 10 Herbst, et. al., JCO, Oct 20, 2007

Tarceva Other Tumor Types Planning Phase III registration study in hepatocellular carcinoma Nexavar + Tarceva vs Nexavar Phase III 830 patient front-line maintenance study in ovarian cancer (EORTC) Over 200 IST and CTEP collaborative programs across multiple disease settings help to pave the way and drive broad publication strategy

Historical Comparison of Tarceva, A+T, and Nexavar in HCC A/T Phase II Data Suggests Anti-EGFR/VEGF (Angiogenesis) Combination Therapy Can Dramatically Improve Survival

Tarceva Launching New Paradigms Chemotherapy paradigm Tarceva can be a better choice than conventional cytotoxic therapy Chronic maintenance paradigm Tarceva po QD enables cancer treatment to be chronic maintenance therapy for remission prolongation, secondary prevention, & possibly primary prevention Target-driven paradigm Understanding of Tarceva MOA allows for rational clinical development & patient selection, rather than tumor-focused fishing expeditions Molecular cocktail paradigm Tarceva can become the new "anchor drug" for non-cytotoxic cocktails, guided by a molecular targeting approach Treatment beyond progression paradigm Tarceva remains a treatment component even when single agent fails

Drug Discovery & Translational Research Focused on EMT David M. Epstein, Ph.D. Senior Vice President, OSI Oncology

OSI Oncology R&D EMT Platform: Leadership in Emerging Field Our Oncology Thesis: "Delivering an Optimized Drug for a Molecular Target to Selected Patients in the Best Biologically Driven Combination With Other Molecular Targeted Therapies Will be Essential to Delivering the Substantial Efficacy Improvements That Will be Necessary to Command Premium Pricing and Drive Market Share in the Increasingly Crowded and Price Sensitive Oncology Arena of the Future."

Epithelial-to-Mesenchymal Transition Research & Development Thesis Focusing on EMT provides OSI the opportunity to Target unique tumor processes Design rational molecular targeted therapies Develop rational MTT combinations Link targets to molecular markers of progression ? Patient selection markers Optimize Drugs to a Molecular Target that drives a Unique Tumor Process & Acts in Combination with other MTTs

Epithelial-to-Mesenchymal Transition Research & Development Thesis Topics, Issues, How we are thinking about EMT What is EMT and how do we target it? Tumor heterogeneity and multi-targeting Alignment of our current portfolio with the OSI EMT thesis AVEO collaboration

Epithelial-to-Mesenchymal Transition Essential for Normal Development EMT is an essential cellular program for embryonic development Allows for mobilization of epithelial cells and movement to new sites to form organs and tissue In normal EMT, Epithelial cells de-differentiate to form Mesenchymal cells Characterized by loss of cell adhesion, increased mobility In normal EMT, Mesenchymal cells migrate to a distal site and re-differentiate into Epithelial cells Characterized by gain of cell adhesion ? organ formation

Pathological EMT Linked to Solid Tumor Progression [A] Most solid tumors start in a differentiated, Epithelial phenotype [B] As tumors progress external (stromal / tumor derived factors) and internal signaling events trigger a transition to a Mesenchymal-cell like phenotype [C] Mesenchymal-like tumor cells invade neighboring tissue, migrate via the vasculature and metastasize at distal sites EMT Blood Vessel Endothelial Cells A B C

EMT-focus evolved from need to understand patient response to Tarceva Tarceva is one of the most selective TK inhibitors developed to date EGFR signaling is thought to be an important driver of NSCLC, pancreatic cancer, GBM and other solid tumor types Tarceva Tarceva and EMT From Bench to Bedside and Back Again

Tarceva-sensitive NSCLC cell lines express Epithelial markers (E-Cad) Tarceva-insensitive NSCLC cell lines express Mesenchymal Markers* Data applies to pancreatic cancer and other tumor types ** Data supports concept that proteins important in EMT may be biomarkers for Tarceva E - - cadherin Fibronectin GAPDH Vimentin H1703 SW1573 H460 Calu6 A549 H441 H358 H322 H292 Epithelial Mesenchymal EGFR E - - cadherin ? - -catenin Fibronectin GAPDH Vimentin H292 Increasing Sensitivity EGFR Tarceva Biomarkers & EMT Identifying Markers That Predict Sensitivity * Thomson et al., Can Res 65: 9455, 2005 ** Buck et al., Mol Can Ther 5: 2051, 2007 Buck et al., Mol Can Ther 6: 532, 2007

87 patients 74% of patients are E-cadherin positive E-cadherin positive patients have a longer time to progression comparing Tarceva+Chemo with Chemo alone treatment. No difference for response or survival | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Progression-Free Rate Chemo Alone, E-cadherin pos Erlotinib+Chemo, E-cadherin pos Chemo Alone, All Patients Erlotinib+Chemo, All Patients Weeks 0 20 40 60 80 0.0 0.2 0.4 0.6 0.8 1.0 HR=0.37 p=0.0028 Clinical Correlates of EMT and Sensitivity E-cadherin Analysis in TRIBUTE Yauch et al., Clinical Cancer Res 15: 8686, 2005.

* Denotes no detected signal PANC1 EGF (10ng/ml) erlotinib (10mM) BxPC3 + + + GAPDH + + + pHER3(Y1289) + + + HER3 + + + pEGFR(Y1068) pAkt(473) + + + pS6(235/236) + + + mesenchymal epithelial EGFR•HER3 AKT S6 proliferation/survival cytoskeletal remodeling erlotinib * * Adapted from Buck et al., Mol. Cancer Ther. 5: 2051-2059, 2006 Tarceva Biomarkers Her3 Is a Key Mediator of Akt.

EMT and Cell Signaling The Basis for New Drug Combinations EGFR Akt MAPK IGF1R PI3K Ras ZEB1, Snail, Twist Integrins Cox2 Ron/Met PDGFR TGFbR Gene transcription ILK b-Cat Epithelial Ph. Mesenchymal Ph. EMT Gene Transcription Akt MAPK EGFR Del mut Her3 IGF1R PI3K Ras a-Cat b-Cat g-Cat Erlotinib a-Cat g-Cat b-Cat E-cadherin a-Cat b-Cat g-Cat a-Cat g-Cat b-Cat Mesenchymal Ph. Epithelial Ph. MET Rapa MMPs PDK-1 PDK-1 m-TOR m-TOR Src OSI-906 OSI-027

EMT Drug Discovery Strategy Focus and Execution Cell Models & Screens Tumor Growth & EMT Targets Drug Discovery To Block EMT, Tumor Growth EMT Tumor Models EMT & Target- Directed Biomarkers Novel Agents & Drug Combinations

Pathological EMT and Tarceva Characterizing Responsiveness In Vitro H1703 Calu6 H441 H292 E-Cadherin Vimentin sensitive refractory Adapted from Thomson et al., Cancer Res., 2005 E-cadherin and vimentin are robust markers to differentiate epithelial and mesenchymal tumor cells

EMT Response Biomarker Correlation In Vitro ? In Vivo Models ? Clinical Samples High E-cadherin expression Low E-cadherin expression E-cadherin is expressed but mutated and non-functional Established Tumor Cell Lines Human Tumors Primary Xenografts Conclusion: Resected pancreatic tumors are E-cadherin positive and sensitive to erlotinib Conclusion: Earlier stage (resectable) pancreatic tumors are epithelial and likely to show enhanced erlotinib-sensitivity Conclusion: EMT status is predictive of erlotinib sensitivity for a panel of established pancreatic and colorectal tumor cell lines OSI-Repasky Lab Collaboration

EMT and Tumor Heterogeneity The Basis For Seeking New Drug Combinations Tumor center is epithelial; invasive front is mesenchymal Tarceva might be expected to have greater activity on the epithelial portion of the tumor and less on the mesenchymal region Heterogeneity ? Multi-targeting OSI EMT biomarker focus: Gain expertise in patient selection and efficacy markers for our MTTs Need new in vivo models that better reflect the growth of tumors in humans (OSI-AVEO Collaboration)

EMT and Cell Signaling Current Portfolio EGFR Akt MAPK IGF1R PI3K Ras ZEB1, Snail, Twist Integrins Cox2 Ron/Met PDGFR TGFbR Gene transcription ILK b-Cat Epithelial Ph. Mesenchymal Ph. EMT Gene Transcription Akt MAPK EGFR Del mut Her3 IGF1R PI3K Ras a-Cat b-Cat g-Cat Erlotinib a-Cat g-Cat b-Cat E-cadherin a-Cat b-Cat g-Cat a-Cat g-Cat b-Cat Mesenchymal Ph. Epithelial Ph. MET Rapa MMPs PDK-1 PDK-1 m-TOR m-TOR Src OSI-906 OSI-027

Alternate RTKs (IGF1R) EGFR EGFR Raf Erk erlotinib Proliferation / Cell Cycle Progression PI3K Akt mTOR S6 PDK1 OSI-906 EGFR HER3 erlotinib Exploiting EGFR-TKI Mediated Resistance Small Molecule-TKI Blockade of IGF-1R

EMT and Cell Signaling Preclinical Profile of IGF-1R Inhibitor (OSI-906) A. 0 4 8 12 16 20 24 0 2 4 6 8 10 0 Time (h) Plasma drug pIGF - - IR Plasma Drug (mM) 0 4 8 12 16 24 2 4 6 8 20 40 60 80 100 120 % pIGF-IR content 0 100 200 300 400 500 600 700 800 900 1 11 21 31 41 51 Day % Tumor Volume Control 25 mg/kg 50 mg/kg 100 mg/kg Dosing period OSI-906 Pharmacodynamic Activity Potent inhibition of IGF-1R activation Durable blockade of pIGF-1R level Dose-dependent Time-dependent OSI-906 Anti-tumor Activity Single agent in vivo activity Dose-dependent

In NSCLC and pancreatic cancer EGFR and IGF-1R combinations show greatest synergy in cells with a competent EGFR-Akt signaling axis Correlates with an ability to reduce phosphorylation of Akt Such a reduction in Akt is not observed in E-cadherin negative cells EMT-based Drug Combinations EGFR and IGF-1R Inhibitors 0 200 400 600 800 1000 1200 1400 1600 1800 1 21 41 61 81 Day Median Tumor Volume (mm3) Control T T + I (3/8 cures) I H292 NSCLC BxPC-3 Pancreatic xenografts 0 100 200 300 400 500 600 1 6 11 16 Day % Tumor Volume Control I T T + I

Exploiting EMT & Resistance Small Molecule-TKI Blockade of TORC1/TORC2 EGFR EGFR growth factor receptors (IGF1R / c-Met / PDGFR) PI3K Akt Raf mTORC2 S6 Erk erlotinib Proliferation / Survival / Cytoskelatal Remodeling Constitutive activity: Mutations of PI3K loss of PTEN PDK1 mTORC1 OSI-027 Rapamycin OSI-027

EMT and Cell Signaling The Basis For New Discovery Programs EGFR Akt MAPK IGF1R PI3K Ras ZEB1, Snail, Twist Integrins Cox2 Ron/Met PDGFR TGFbR Gene transcription ILK b-Cat Epithelial Ph. Mesenchymal Ph. EMT Gene Transcription Akt MAPK EGFR Del mut Her3 IGF1R PI3K Ras a-Cat b-Cat g-Cat Erlotinib a-Cat g-Cat b-Cat E-cadherin a-Cat b-Cat g-Cat a-Cat g-Cat b-Cat Mesenchymal Ph. Epithelial Ph. MET Rapa MMPs PDK-1 PDK-1 m-TOR m-TOR Src OSI-906 OSI-027

OSI-AVEO Collaboration Discovery and Translational Research Discovery and Translational Research Discovery and Translational Research Discovery and Translational Research Discovery and Translational Research AVEO Tumor Model Platform Target-driven tumor models Discovery programs EMT process linked to tumor growth in vivo Tumor archives for Biomarkers Patient selection Drug response IGF-1R Response Models: OSI-906 mTOR Response Models: OSI-027

EMT Drug Discovery Strategy Focus and Execution OSI EMT Discovery Platform Drivers of tumor growth & EMT e.g., RON/cMET, FAK, SRC EMT model systems for drug discovery OSI IP and Technology Development Academic & AVEO collaborations Pharmacological Disruption of EMT Demonstrated and Feasible Yields sustained sensitivity to TKIs EMT and Tumor Heterogeneity Blockade of E- & M-cell pathways Synergy among multiple MTTs Unique oncology strategy based on rational MTT combinations

Summary Focus on EMT & Drug Discovery Epithelial tumor cells use EGFR-mediated growth / survival pathways EMT contributes to de novo and acquired resistance to EGFR-TKIs Redundant pathways limit sustained response to EGFR-TKIs Synergistic MTT development (IGF-1R & TORC1/TORC2)

Summary of OSI EMT Research Building An EMT-Focused Discovery Engine Pioneered understanding of EMT and response to EGFR-TKIs Developing EMT models to understand EGFR TKI resistance, drug combinations, and new EMT-targeted drugs Identified and validated EMT targets for drug development such as RON/MET, FAK/SRC, IGF-1R, TORC1/TORC2 Identifying and developing combinations to enhance MTT activity OSI-AVEO Developing EMT-focused discovery programs (+ IP) Clinical biomarkers and response platform (+ IP)

Development Programs Karsten Witt, M.D. Vice President, Clinical Development, OSI Oncology

OSI-906

IGF-1R as an Anti-Cancer Target Rationale IGF-1R widely expressed among epithelial- derived tumors Receptor for both IGF-1 (liver) and IGF-2 (autocrine/paracrine) Tumor xenograft studies with IGF-1R kinase inhibitors and anti-IGF-1R antibodies indicate: Tumors with autocrine overexpression of IGF-2 ligand are dependent on IGF-1R Colorectal, ovarian, pediatric sarcomas Sensitive to IGF-1R inhibitors as monotherapy Tumors with high IGF-1R expression but low autocrine IGF NSCLC, breast, prostate Partial sensitivity to IGF-1R inhibition May require IGF-1R inhibitor in combination with other agents (e.g. Tarceva) for maximal anti-tumor efficacy

IGF-1R Inhibitor OSI-906 Proof of Concept Potent compound identified that has 10-50X selectivity versus IR Excellent anti-tumor activity observed in mouse models of human tumor xenografts driven by IGF/IGFR Good synergy in defined tumors with Tarceva Hyperglycemia observed in animal models but only at exposures 10X those required for efficacy IND/CTA filed in Dec 06 for OSI-906 0 200 400 600 800 1000 1200 1400 1 11 21 31 Day % Tumor Volume Control 25 mg/kg 75 mg/kg

OSI-906 Differentiation Strategy First-in-Class oral IGF-1R TKI Potentially more effective than antibodies given its inhibition of the pAKT survival pathway Convenience and flexibility of oral administration Shorter half-life Improved ability to manage adverse effects Short-term discontinuation for toxicity management if needed Ability to sequence with other agents Accelerated development in combination with Tarceva

OSI-906-101Accelerated Phase I Program Continuous Daily Dosing Phase I Dose Escalation initiated June 07 n = 3+3 At 'trigger' eg, gr. 2 tox or PK etc, open BID QD Dosing Escalation with expansion at MTD BID Dose Escalation with expansion at RD Starting Dose 10 mg QD 28-Day Treatment Cycle 28-Day Treatment Cycle Extensive PK, safety monitoring, home glucometers 20 mg QD dose enrolling with no safety concerns

OSI-906-102 Accelerated Phase I Program Intermittent Daily Dosing At 'trigger' eg, gr. 2 tox or PK etc, open S2 followed by S3 QD Day 1-3 Dosing Escalation with expansion at MTD Starting Dose 10 mg QD 14-Day Treatment Cycle 14-Day Treatment Cycle Designed to integrate with FOLFOX or other chemoTx 40 mg QD dose currently enrolling with no safety concerns S1 S2 S3 Phase I Dose Escalation initiated July 07 n = 3+3

OSI-906 2008 Deliverables Ongoing research In-vivo combination studies with MTT and chemotherapy Optimization of PD biomarker assays Identification of potential patient selection markers Progress Phase I studies toward recommended dose 906-101: 2 schedule cohorts (Daily QD and BID dosing on a 28 day cycle) 906-102: 3 cohorts (Days 1-3, 1-5, and 1-7 on a 14 day cycle) Be ready to initiate Phase Ib and Phase II programs seamlessly upon completion of Phase I studies Initiate combination study with Tarceva as soon as possible

OSI-906-201 Phase I/II Tarceva Combination Tarceva + OSI-906 N ~ 60 Tarceva + placebo N ~ 60 Phase I Dose Escalation Starting at Tarceva 100 mg/d OSI-906 X mg/d N = ~ 25 R PII D Randomize Ongoing Exploration of Potential for a Patient Selection Strategy AVEO Biomarker Collaboration

Summary of OSI-906 Clinical Development Plan Strategic/Tactical positioning First-in-Class oral IGF-1R inhibitor, strong IP position, "hot" target Differentiation First small molecule to clinic, flexibility to dose with other agents, short half-life vs MAb Potential for biomarker driven patient selection "Step-up" value Tarceva + OSI-906 combination program accelerated via Phase I/II design Synergistic activity seen in preclinical models Opportunities for both speed to market and large market development strategies

OSI-027

mTOR Kinase Inhibitor Project Improved Approach to a Proven Key Target mTOR is a proven pharmacologic target: mTOR signaling is frequently activated (regulated) by mutations (PI-3K) and inactivation (PTEN) Rapamycin analogs (eg. temsirolimus & everolimus) inhibit the function of the mTOR-Raptor complex OSI mTOR Kinase inhibitor are next generation modulators of mTOR signaling Inhibit both pS6K (TORC1) and pS473-Akt (TORC2) signals in tumor cells as well as 4EBP1 phosphorylation Effectively inhibit proliferation of rapamycin sensitive and insensitive cell lines in culture Inhibit growth of rapamycin-insensitive tumors in vivo after oral administration

OSI-027 Status Update Enabling toxicology studies to prepare regulatory clinical trial submissions completed GMP API completed and prototype drug product formulations on stability Report writing ongoing Phase I protocol in final draft stage EU CTA/IMPD targeted for end of 2007 US IND to follow in 2008

Phase I Dose Escalation N = ~ 15 OSI-027-101 Phase I: Cyclic and Continuous Dosing Daily Dosing (dilution of 1-3d Q7) N < 20 with expansion at RD S3 At 'trigger' eg, g2 tox or PK etc, open S2 and S3 S2 1d Q7 Dosing Escalation N < 20 with expansion at RD 1-3d Q7 Dosing Escalation N < 20 with expansion at RD S1 1-3d Q7 Dose Escalation Dosing of first patient estimated by 2Q08

OSI-027 Dose Scheduling Study in IGROV1 Tumor Model QD, Q2D, Day 1-14 Q4D, Day 1-14 QD, Q2D, Q3D, Q4D, Q7D, Day 1-14

OSI-027 Differentiation Strategy Torc1 and Torc2 inhibition vs Torc1 only for rapalogues Orally available small molecule with predictable PK/exposure compared to rapalogues Active in many tumor cell lines and xenografts: rapalogue sensitive and resistant Good opportunity for first and Best-in-Class

OSI-930

Dual Kit/KDR Inhibitor Rationale Kit involvement in cancer Kit mutation and activation in majority of GIST patients Kit mutations in other tumor types (mast cell leukemia, seminoma, AML) Kit expression in small cell lung cancer (-/+ Kit ligand) KDR involvement in cancer KDR-dependant angiogenesis required for continued growth of most solid tumor types N N H S O N H O C F 3 N N H S O N H O C F 3 O OSI-930 OSI-817

Dual Kit/KDR Inhibitor Current Status OSI-930 Healthy volunteer Phase I study completed (OSI-930-101) PK linear through 300 mg 63 subjects (single dose) 0-300 mg Phase I studies in cancer patients ongoing Single-agent QD and BID dose escalation ongoing (OSI-930-102) Combination with Tarceva initiated (OSI-930-103) Molecular markers being evaluated pS6 a good downstream marker Phase II strategies could include combination with Tarceva in NSCLC and/or targeting a Kit-dependent tumor type (e.g. SCLC)

OSI-930-102 Phase I trial designed to explore QD and BID schedules 27 patients enrolled to date 12 patients QD schedule (50 - 1600 mg) Minimal toxicity Further escalation limited by number of tablets 15 patients BID (400 - 600 mg) DLT in 3 of 8 patients (1 gr. 4 GGT, 2 gr.3 Rash) at 600 mg PK at 600 mg BID suggests trough levels ~2.74 µM, above target range from preclinical 500 mg intermediate dose-level currently enrolling Evidence of anti-tumor activity across both schedules Best response: 1 PR (ovarian); 10 SD; 8 PD; 4 too early DCR = 50% in QD cohort and 70% in BID cohort Phase I Single-Agent Trial in Cancer Patients

Complete Phase I program Explore evidence for Kit driven activity (SCLC, 3rd-line GIST) Assess overall data package Dose escalation study in combination with Tarceva Determine if clear differentiation strategy exists while exploring a broad based program with a global partner OSI-930: Summary Plans for 2008

Oncology Q&A

Overview of Diabetes/Obesity Programs Anker Lundemose, M.D., Ph.D. President, (OSI) Prosidion

Value Drivers of OSIP's Diabetes & Obesity Franchise DP-IV Patents Estate, Licenses & Revenues PSN821: Novel oral activator of one of the most exciting targets in type 2 diabetes today, with both potential anti-diabetic and appetite suppressing features - entering Phase I Q3 2008 PSN602: Novel obesity approach with potentially superior efficacy and designed to confer cardiovascular neutrality - entering Phase I by Q3 2008 PSN010: Potentially superior efficacy from a novel anti-diabetic class (Glucokinase Activator) which is in the hands of the world's leading diabetes company Eli Lilly - currently in Phase I Strong discovery engine in Diabetes & Obesity

DP-IV Patent Estate & License Portfolio Strong Potential for Revenue & Cash Flow Ownership/control of patent estate for the use of DP-IV inhibitors for the treatment of Diabetes, as monotherapy and in combination Acquired as part of $35MM acquisition of DP-IV assets from Probiodrug in 2004 Additional value added to the patent estate over the last 3 years 12 large Pharma companies have signed license agreements including Merck, Novartis, BMS (Astra-Zeneca) and Roche $62.5 million in upfront milestones and royalties generated to date (excluding potential Q4 royalties)

DP-IV Patent Estate & License Portfolio Januvia Off To A Head Start I:\Active\OSP17\Research\[DiabetesMarketResearch_JM_v2.xls]For Slides 2003 04 05 06F 07F 08F 09F 10F Sulfonylureas 0.59 0.46 0.48 0.22 0.18 0.17 0.16 0.15 Biguanides 1.22 0.53 0.71 0.68 0.7 0.71 0.72 0.75 Meglitinides and Amino Acid Derivatives 0.18 0.19 0.21 0.2 0.15 0.13 0.14 0.14 Insulin Sensitizers (incl. next gen PPARs) 2.53 2.95 3.5 3.98 4.39 4.72 4.99 5.19 DPP-IV Inhibitors 0 0 0 0 0.53 1.5 2.39 3.16 Billions of Dollars U.S. OAD Market Size (2003 - 2010F) Other OADs* Insulin Sensitizers DP-IV inhibitors CAGR% 16.3% 1.8% Total 4.0% 03-06F 07F-10F 5.7% 1.0% 16.4% The launch of Januvia, the first DP-IV inhibitor, has exceeded expectations The issues relating to Novartis' Galvus have given Merck's Januvia a significant advantage Januvia sales totaled $459 million in first 9 months of 2007 Annual royalty stream from DP-IV patent estate would approximate $100 million for each $5 billion of sales for DP-IV market Sulfonylureas Metformin (17.4%) 2.5% (27.8%) (7.5%) 81.7% Note : *Other OADs includes prandial glucose regulators and alpha-glucosidase inhibitors Source : Credit Suisse First Boston, L.E.K. analysis

Type 2 Diabetes is a Progressive Disease Current Treatment Paradigm Years from diagnosis beta-cell function (%) 100 0 Prediabetes Diet & Exercise Monotherapy Combination Therapy Insulin 60 40 20 80 - -5 - -4 - -1 - -2 - -3 6 5 0 1 2 3 4 Type 2 diabetes } - - Metformin - - PPARg agonists KATP channel closers DP-IV inhibitors/GLP-1 Despite the introduction of several new drug classes over the last decade there is still a huge unmet need - HbA1c still too high! Current treatments often associated with either gastrointestinal side- effects, nausea, weight gain, hypoglycemia and/or potential cardiovascular effects

Diabetes & Obesity - Major Unmet Medical Needs - Our View Effect on Type 2 diabetes disease progression Improved glucose lowering efficacy & improved side effect (S/E) profile Glucose lowering greater than combination of two current agents, with acceptable side-effect profile ^-cell secretagogue - same or better efficacy as sulfonylurea with better S/E profile Insulin sensitizer - same efficacy as TZDs, but without weight gain & oedema Weight loss Weight loss of 10% or more without major S/E Glucose lowering at least as good as DP-IV inhibitors, with acceptable side-effect profile and meaningful weight loss

OSIP Therapeutic Targets for Type 2 Diabetes & Obesity CNS Peripheral Tissues Gut, Liver & Pancreas SUs Metformin Acarbose(r) Avandia Actos Insulin Xenical(r) Meridia(r) Acomplia(r) Byetta(r) Januvia Beta-cell targets: GKA (PSN010), GPR119 (PSN821) Appetite target: SNRI & 5-HT1A (PSN602) Satiety Gut Target GPR119 (PSN821) New Insulin Resistance Targets Liver Target GKA (PSN010)

Update on PSN821 Jim McCormack, Ph.D, D.Sc. Senior Vice President, (OSI) Prosidion

Gut, Liver & Pancreas PSN821 is a Novel GPR119 Agonist One of the Most Exciting Targets in Type 2 Diabetes Today GPR119 GPR119 Receptor expressed in gut and pancreas PSN821 - Novel oral class of drug with potential anti-diabetic and appetite suppressing features PSN821 releases endogenous GLP-1 and increases beta-cell cAMP leading to improved glucose homeostasis, delayed gastric emptying and appetite suppression

SLOWED DIABETES PROGRESSION WEIGHT LOSS Pancreas GPR119 GLP1R IMPROVED GLUCOSE HANDLING GPR119 GIPR GLP1R cAMP ^ Glucose-stimulated Insulin Secretion ^ b-CELL PROTECTION GLP1 ^ GIP ^ Satiety ^ Food Intake ^ Gastric Emptying ^ Glucagon secretion ^ GPR119 ? Proposed Mechanisms for GPR119 Agonists Action An Exciting New Approach to Diabetes Therapy

GPR119 Localizes to Pancreatic Islets Subset of cells in the islets of Langerhans with strong GPR119 staining Pancreatic exocrine acinar epithelium (negative) Red = GPR119 (alkaline phosphatase chromogen) Blue = nuclear counterstain (hematoxylin) Islet of Langerhans and acini from normal pancreas obtained at autopsy from a 85 years-old female Islets from normal pancreas obtained at surgery from a 76 years-old male Preliminary immunocytochemistry data in human islets using GPR119 polyclonal antibody

GPR119 strong staining in absorptive epithelium within intraepithelial neuroendocrine cells GPR119 Localizes to Endocrine Cells within the Intestinal Villi Longitudinal section of normal small intestine obtained at surgery from a 69 years-old female Red = GPR119 (alkaline phosphatase chromogen) Blue = nuclear counterstain (hematoxylin) Preliminary immunocytochemistry data in human intestine

Insulinotropic & GLP-1otropic Actions in Cells Insulin secretion in HIT-T15 cells - -3 - -2 - -1 0 1.0 1.2 1.4 1.6 1.8 2.0 2.2 Log[PSN821] (mM) Insulin secrxn (fold) EC50 = 153 nM Emax = 2.4 Log[PSN821] (mM) - -3 - -2 - -1 0 1 GLP-1 secrxn (fold) 1.0 1.2 1.4 1.6 1.8 2.0 2.2 GLP-1 secretion in GLUTag cells EC50 = 130 nM Emax = 2.0

GLP-1 Secretion in SD Rats 0 5 10 15 20 25 30 35 Active GLP-1 (pM) T = 0 min T = 45 min ** 0 10 20 30 ** D[GLP-1]0-45min DP-IV Inhibitor + PSN821

3 g/kg Glc Oral Glucose Tolerance Test in Mice PSN821 DP-IV Inhibitor - -45 0 45 90 135 180 6 8 10 12 14 16 Time (min) [Glucose] (mM) ** ** ** ** * * 0 100 200 300 400 Vehicle DP-IV i PSN821 *** *** Reactive Glc AUC (mM ? 120 min) Cmpd

Inhibition of Gastric Emptying in SD Rats Vehicle (20% HP-b-CD) Atropine 0 10 20 30 40 50 60 70 80 90 * * * % Gastric content @ T = 20 min PSN821 Dose 2 PSN821 Dose 3 PSN821

Diabetes Progression in Prediabetic ZDF Rats N.B. Rat iv T1/2s: PSN821: 1.8 h; Vildagliptin: 8.8 h 5 10 15 20 - -5 0 5 10 15 20 25 30 35 40 45 50 Treatment day Fed blood [Glc] (mM) Vehicle Low dose PSN821 High Dose PSN821 10 mg/kg/d Vildagliptin * * * * * * * * * *

Body Weight Loss in Female DIO Rats 370 380 390 400 410 420 430 440 450 - -6 - -5 - -4 - -3 - -2 - -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Day Body weight (g) Vehicle 5 ml/kg po PSN821 low dose po PSN821 high dose po PSN699540 high dose po Vildagliptin 10 mg/kg po Sibutramine 5 mg/kg po - -2.2% - -1.8% - -10.6% - -6.6% - -8.8%

Vehicle (20% HP-b-CD) 10 mg/kg/d Rimonabant PSN821 DP-IV inhibitor Retroperitoneal Epididymal Total 0.0 0.5 1.0 1.5 2.0 2.5 ** ** * ** * Tissue weight (g) Fat Pads 0 10 20 30 [Leptin] (ng/mL) Vehicle ** Rimon- abant ** PSN821 DP-IV inhibitor Leptin 0.0 0.2 0.4 0.6 0.8 Vehicle DP-IV inhibitor Rimonabant * PSN821 [Triglyceride] (mM) TGs Fat Pad/Leptin/TG Lowering After 28d Dosing to DIO Mice

Update on PSN602 & PSN010 Jonathan Rachman, M.D., Ph.D. Senior Vice President, (OSI) Prosidion

PSN602

PSN602 - A Novel Obesity Approach Dual monoamine reuptake inhibitor & 5-HT1A agonist for the treatment of obesity Noradrenaline and serotonin reuptake inhibition ^ appetite reduction Increased noradrenaline activity associated with increases in BP and heart rate; 5-HT1A agonism included to potentially counter-balance undesirable CV effects ? Improved safety profile ? Improved efficacy

Sibutramine - Phase IIa Study Approved doses Greater efficacy but not approved due to CV effects Bray et al, Obesity Research 1999, Vol. 7, No. 2, p 193 Sibutramine is associated with dose-related increases in HR and BP which limits its dosing and hence its efficacy

PSN602 - Efficacy in Female DIO* Rats *Chronically obese due to high fat feeding for months: "gold standard" obesity model

PSN602 - No Significant Preclinical CV Effects At up to 3xED50 24h in acute feeding - -2 0 2 4 6 8 10 12 6.9 mg/kg 13.8 mg/kg PSN602 po 20.7 mg/kg Sibutramine 15 mg/kg po Mean arterial blood pressure (difference from vehicle) *** BP - -20 - -10 0 10 20 30 40 50 60 70 6.9 mg/kg 13.8 mg/kg 20.7 mg/kg Sibutramine 15 mg/kg po Heart rate (difference from vehicle) * *** PSN602 po HR Mean arterial BP (mmHg) & HR (bpm) (mean 2 to 8 hours after dosing, D from vehicle)

PSN602- Targeted Design Provides Unique Positioning 5-HT1A agonist PSN602 Edronax(r) (Pfizer) Strattera(r) (Lilly) (Depression, ADHD) e.g. Prozac(r) (Lilly) (Depression) Buspar(r) (BMS) PRX-00023 (EPIX) (Anxiolytics) SRI NRI DRI Tesofensine Sibutramine PSN Chemistry Vilazodone

Tesofensine - A Triple Reuptake Inhibitor 24-week dosing Phase IIb study "AEs were mostly mild to moderate: dry mouth, sleep disturbances, nausea, constipation and diarrhea" "Slight increases in heart rate and blood pressure" TIPO-1 results Placebo Tesofensine 0.25 mg Tesofensine 0.5 mg Tesofensine 1.0 mg ITT population 52 52 50 49 Mean weight at base line (kg) 103.2 101.7 100.1 101.3 Average relative weight loss 2.0% 6.5%** 11.2%** 12.6%** Average absolute weight loss (kg) 2.2 6.7** 11.3** 12.8** Source: Neurosearch A/S press release September 17 2007

PSN602 - Potential PSN602 delivers superior efficacy relative to sibutramine in DIO rats PSN602 has no dopamine reuptake inhibition element PSN602 is designed to avoid increases in heart rate and blood pressure PSN602 may deliver similar efficacy to that seen with Tesofensine in humans without dopamine and CV related issues

PSN602 - A Novel Anti-Obesity Candidate PSN602 has the potential to deliver both greater efficacy and an improved side-effect profile compared with existing drugs approved for the treatment of obesity PSN602 is currently undergoing scale-up and regulatory preclinical studies Phase I start planned for Q3 2008 Data expected by end Q4 2008

PSN010

Gut, Liver & Pancreas PSN010 PSN010: A Novel Glucokinase Activator Dual effects in Pancreas and Liver Increased glucose uptake by the liver Stimulates insulin secretion by the pancreas Positioning: Single agent with dual mechanism of action - potentially superior efficacy Program licensed to Eli Lilly in January 2007 Upfront $25MM Milestones & other payments up to $360MM Tiered lower double-digit royalties

(OSI) Prosidion: Closing Remarks Anker Lundemose, M.D., Ph.D. President, (OSI) Prosidion

Effect on Type 2 diabetes disease progression Improved glucose lowering efficacy & Improved side effect (S/E) profile Glucose lowering greater than combination of two current agents, with acceptable side-effect profile ^-cell secretagogue - same or better efficacy as sulfonylurea with better S/E profile Insulin sensitizer - same efficacy as TZDs, but without weight gain & oedema Weight loss Weight loss of 10% or more without major S/E Glucose lowering at least as good as DP-IV inhibitors, with acceptable side-effect profile and meaningful weight loss PSN010 PSN010 PSN602 Key Discovery Focus Key Discovery Focus PSN821 PSN821 PSN821 Diabetes & Obesity - Major Unmet Medical Needs - Positioning of OSIP's Pipeline

Data & Timelines Compound Status Next event Next Key Data expected Nature of Key Data PSN821 Preclinical Phase I initiation Q3 2008 End Q4 2008 Initial Phase I data: Safety, PK and OGTT* PSN602 Preclinical Phase I initiation Q3 2008 End Q4 2008 Initial Phase I data: Safety, PK and appetite * Clinical trial design subject to approval by regulators Phase II Studies and Readouts Anticipated in 2009

Diabetes and Obesity Q & A

Financial Commentary Michael G. Atieh Chief Financial Officer

A Word About Taxes

OSI's Historical Net Operating Losses Represent a Significant "Unrecognized" Asset OSI will have US and UK NOLs of approximately $1 billion at 12/31/07 NOLs represent the opportunity to offset future taxable income on a dollar for dollar basis However, there are limits on the annual use of NOLs Our current estimate of the annual limitation is approximately $225 million per year

OSI's Historical Net Operating Losses Represent a Significant "Unrecognized" Asset OSI will have US and UK NOL's of approximately $1 Billion at 12/31/07 NOL's represent the opportunity to offset future taxable income on a dollar for dollar basis However, there are limits on the annual use of NOL's Our current estimate of the annual limitation is approximately $225 Million per year Any unused limit in one year can be applied to the following year(s) OSI will have a cash tax cost if annual income is in excess of these annual limits (plus any unused balance from previous years)

GAAP Treatment of NOLs OSI's NOLs represent "unrecognized" Deferred Tax Assets Accounting guidelines require OSI to recognize Deferred Tax Assets, provided that NOL "utilization" is deemed likely in a future period For many companies with historical losses, this criteria is typically met in the third year of profitability, assuming future income is projected For OSI, assuming we meet the standard of future income, this will occur in 2009 or as early as 4Q08

GAAP Treatment of NOLs OSI's NOL's represent "unrecognized" Deferred Tax Assets Accounting guidelines require OSI to recognize Deferred Tax Assets, provided that NOL "utilization" is deemed likely in a future period For many companies with historical losses, this criteria is typically met in the third year of profitability, assuming future income is projected For OSI, assuming we meet the standard of future income, this will occur in 2009 or as early as 4Q08 Future Financial Statement Impact: Significant, non-cash credit in P&L for tax value of remaining unused NOLs Full tax provision in P&L based on statutory rates Minimum cash tax liability until NOLs are fully utilized

Financial Commentary

2007: A Year of Financial Transformation Demonstrated the significant earning power of "Core" Business Tarceva as primary engine Strong growth on a global basis DP-IV patent estate as secondary engine Deal flow is slowing but royalties are rising fast Streamlined P&L provided leverage We estimate the Core Business will contribute approximately $1.15 of the $1.65 EPS guidance Lilly deal for PSN010 was "opportunistic" and targeted as part of a key financial transition year for OSI Renovo revenue was unanticipated 2007 Core Business results of $1.15 per share establish a solid "Baseline Year" for OSI

2007: A Year of Financial Transformation Demonstrated the Significant Earning Power of "Core" Business Tarceva as primary Engine Strong growth on a global basis DPIV Patent Estate as secondary Engine Deal flow is slowing but Royalties are rising fast Streamlined P&L provided leverage We estimate the Core Business will contribute approximately $1.15 of the $1.65 EPS guidance* Lilly deal for PSN 010 was "opportunistic" and targeted as part of a key financial transition year for OSI Renovo Revenue was unanticipated 2007 Core Business results establish a solid "Baseline Year" for OSI *note: Guidance as of OSI's 10/25/07 earnings call - OSI is not updating nor reaffirming guidance at this time A key Financial Goal is to generate profitable growth over the next five years from our Core Business compared to the Baseline Year 2007, while balancing the need to invest in R&D

2008: A Year of Scientific Transformation But With Continued Focus on Financial Discipline Focus R&D efforts on technology and projects where we are or can be industry leaders Key goal is to establish a follow-on portfolio with Core Assets that we own and control Spend high percentage of "D" dollars where there is greatest likelihood to create a near term step-up in value for OSI shareholders Develop Non-Core Assets to optimize potential partnering value Objective is to generate revenue from an "Out-Licensing/Partnering Pool" of assets Provide near-term cash flow, earnings leverage, and the potential for a payoff down the road Out-Licensing/Partnering contributed $32M in revenues in 2007 Maintain controls around SG&A spending

2008: A Strong Financial Follow-On Year to 2007 Our Early Assessment We anticipate that global Tarceva sales should grow by >$200 million over 2007 guidance Minimal incremental costs associated with growth DP-IV related royalties may double 2007 results Januvia/Janumet likely to dominate & class may reach $1.5 billion in 2008 based on market research Continued focus on SG&A expenses, coupled with some one- time items in 2007, drives lower spending in 2008 R&D spending modestly higher in 2008 as product candidates move through pipeline

R&D Spend in 2008 Anticipated to be Approximately $140 M

2008: A Strong Financial Follow-On Year to 2007 Our Early Assessment We anticipate that global Tarceva sales should grow by >$200 million over 2007 guidance Marginal costs associated with growth DP-IV related royalties may double 2007 results Januvia/Janumet likely to dominate & class may reach $1.5 billion in 2008 based on market research Continued focus on SG&A expenses, coupled with some one- time items in 2007, drives lower spending in 2008 R&D spending modestly higher in 2008 as product candidates move through pipeline Core Business Should Deliver EPS of ~$2.00 for 2008

2008: A Strong Financial Follow-On Year to 2007 Our Early Assessment We anticipate that global Tarceva sales should grow by >$200 million over 2007 guidance Marginal costs associated with growth DP-IV related royalties may double 2007 results Januvia/Janumet likely to dominate & class may reach $1.5 billion in 2008 based on market research Continued focus on SG&A expenses, coupled with some one- time items in 2007, drives lower spending in 2008 R&D spending modestly higher in 2008 as product candidates move through pipeline Pro-active Business Development efforts to license/partner Non-Core Assets should result in revenues that offset higher R&D spending

Financial Q&A

Wrap-up and Final Q&A

Osi Pharmaceuticals (OSIP) Inactive 8-KRegulation FD Disclosure