Advancing a class on top of statins February 2017

Safe Harbor Statement 2 This presentation includes forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended. Except for statements of historical fact, any information contained in this presentation may be a forward-looking statement that reflects the Company’s current views about future events and are subject to risks, uncertainties, assumptions and changes in circumstances that may cause events or the Company’s actual activities or results to differ significantly from those expressed in any forward-looking statement. In some cases, you can identify forward-looking statements by terminology such as “may”, “will”, “could”, “would”, “should”, “plan”, “predict”, “potential”, “project”, “expect,” “estimate,” “anticipate,” “intend,” “goal,” “strategy,” “believe,” and similar expressions and variations thereof. Forward-looking statements may include statements regarding the Company’s business strategy, market size, potential growth opportunities, capital requirements and use of proceeds, clinical development activities, the timing and results of clinical trials, regulatory submissions, potential regulatory approval and commercialization of the product candidate. Although the Company believes that the expectations reflected in such forward-looking statements are reasonable, the Company cannot guarantee future events, results, actions, levels of activity, performance or achievements. These forward-looking statements are subject to a number of risks, uncertainties and assumptions, including those described under the heading “Risk Factors” in our filings with the SEC. These forward-looking statements speak only as of the date of this presentation and the Company undertakes no obligation to revise or update any forward-looking statements to reflect events or circumstances after the date hereof. This presentation also contains estimates and other statistical data made by independent parties and by us relating to market shares and other data about our industry. This data involves a number of assumptions and limitations, and you are cautioned not to give undue weight to such estimates. The trademarks included herein are the property of the owners thereof and are used for reference purposes only. Such use should not be construed as an endorsement of such products.

Gemphire Corporate Overview 3 A Phase 2 clinical CV/NASH biotech company with first-in-class oral, once-daily gemcabene licensed from Pfizer Safety and efficacy validated based on comprehensive date from 895 subjects across 18 trials (11 Phase 1 and 7 Phase 2) and over 30 preclinical toxicology studies $30M IPO at $10/share in August 2016 (NASDAQ:GEMP) with ‘blue-chip’ biotech investors Cormorant, Adage, Excel Venture Mgt, Capital Midwest, Pfizer, and Management/Directors Analyst coverage from Jefferies, RBC, Canaccord Genuity, Laidlaw, and LifeSci Capital Large, unmet global market opportunity despite new therapies Gemcabene differentiated MOA (cholesterol, triglycerides, & inflammation) in cardiovascular disease with upside potential for the treatment of NASH (e.g., recent POC data, AZURE-1 Phase 2 planned) Significant clinical catalysts expected in 2017 Data readouts expected in 3 Phase 2b trials COBALT-1, ROYAL-1, INDIGO-1 in dyslipidemic patients Established regulatory approval pathway with surrogate lipid (LDL-C or TG) endpoints for the selected patient populations Leadership team with track record developing/commercializing/partnering multiple CV and orphan drugs based in Michigan (the origins of best selling drug Lipitor) Valuation at ~$100M (as of 2/3/17), GEMP CV and NASH peer group may include MDCO, ESPR, AMRN, MDGL, ICPT, CNAT, TBRA (acquired by Allergan), ENTA, and GNFT Cash $28.4M (9/30/16), no debt, funding for trials though EoP2 meetings in 1H 2018

NASH Market ($35B+ by 2025) (6) DIABESITY Dyslipidemia Market ($26B) All Statins ~$6.1B(2)(3), Ezetimibe ~$2.7B(2)(3), Bile Acid Sequestrants ~$0.7B(2), PCSK9s ~$0.5B(4), Fibrates ~$1.0B(2), Niacin ~$1.0B(2), Rx Fish Oil ~$1.2B(2) Related Multiple Large Markets of Dyslipidemia and NASH An Estimated 20 Million US Patients Addressable by Gemcabene 4 Cholesterol Inflammation Triglycerides ~10M (1) ASCVD Patients Mixed Dyslipidemia Non-Familial Hypercholesterolemia ~1M (1) FH Patients HeFH HoFH ~3M (1) SHTG Patients ~6M (5) NASH Patients (NAFLD estimated over 60M) Company estimate based on DRG Market Data, NHANES and FH Foundation. Estimated revenues for U.S. dyslipidemia market sales in 2013. Includes revenues when used as a standalone therapy and as a fixed-dose combination therapy. Projected revenues for U.S. sales in 2016. The National Institute of Diabetes and Digestive and Kidney Diseases, 2016 Estimated peak global market for NASH treatments, Deutsche Bank 2014

Note: Percentages (Mean and Median - LDL-C, Median - hsCRP, TG) have been demonstrated across multiple clinical studies. A decrease in CRP (an earlier clinical assay) correlates with a decrease in hsCRP (the current CV risk assay). A NOVEL, COST-EFFECTIVE, ONCE-DAILY ORAL THERAPY Additive Effect in Combination w/ Statins Lowered LDL-C ~30% Promising Safety and Tolerability No myalgia as monotherapy No liver toxicities No significant effect on kidney function No QTc prolongation No clinically meaningful change in blood pressure No food effect Significant Lipid-Lowering LDL-C ~30% TG ~40% hsCRP ~40% No Drug-Drug Interactions High dose atorvastatin High dose simvastatin Digoxin Gemcabene Differentiated Product Profile Multiple Important Cardio/Metabolic Benefits Observed in Patients 5 Safety Efficacy

Gemcabene Development Program Overview 895 Subjects Treated with Gemcabene Across 18 Clinical Trials 6 Low HDL-C and Normal or Elevated TG Healthy Obese Non-Diabetic Hypertension 300, 900 mg (add-on various low, moderate and high intensity statins) 300, 600, 900 mg (concurrent 10, 40, 80 mg atorvastatin) 150, 300, 600, 900 mg 900 mg 900 mg (arm with quinapril 20 mg) 150, 450, 900 mg (arm with rofecoxib 25 mg) Patient Population: Doses: Trial: 1027-018 n=66 (GEM=42) 1027-014 n=53 (GEM=26) A4141001 n=277 (GEM=208) 1027-012 n=102 (GEM=43) 1027-004 n=161 (GEM=129) A4141004 n=404 (GEM=242) Hypercholesterolemia 12 wks 8 wks 8 wks 4 wks 12 wks 4 wks 4 wks Completed 11 Phase I Trials in Healthy Volunteers 25 to 1,500 mg 1027-001, -002, -003, -009, -010, -011; A4141003, -005, -006 n=163 (GEM=142) Single and Multiple Dose Studies; Up to 4 wks Osteoarthritis Phase 2 Phase 1 Completed over 30 nonclinical GLP toxicology studies, including: 26-week study in rats and monkeys; 52-week study in monkeys Ongoing two year carcinogenicity studies in mice and rats Preclinical Duration: 900 mg (DDI Study with 80 mg simvastatin) 15 days 300, 900 mg (DDI Study with 80 mg atorvastatin) 1027-015 n=23 (GEM=23) A4141002 n=20 (GEM=20) 1027-008 n=20 (GEM=20) 900 mg 22 days

Indication Phase 1 Phase 2a Phase 2b Phase 3 NDA Anticipated Milestones Homozygous Familial Hypercholesterolemia (HoFH) COBALT-1 Phase 2b trial (n=8) ongoing and interim data provided January 30, 2017; top-line data expected in June 2017 Hypercholesterolemia – Heterozygous Familial Hypercholesterolemia (HeFH) ROYAL-1 Phase 2b trial (n=104) ongoing and top-line data expected in 3Q 2017 Hypercholesterolemia – Atherosclerotic Cardiovascular Disease (ASCVD) Severe Hypertriglyceridemia (SHTG) INDIGO-1 Phase 2b trial (n=90) ongoing and top-line data expected in 4Q 2017 Non-alcoholic Steatohepatitis (NASH) / Non-alcoholic Fatty Liver Disease (NAFLD) AZURE-1 Phase 2 trial protocol being finalized with plans to initiate AZURE-1 in 2017 Gemcabene Pipeline and Clinical Plans Multiple Clinical Catalysts in 2017 7

Rationale for Potential Meaningful Results in COBALT-1 8 COBALT-1 Trial Design (GEM-201) Phase 2b Clinical Trial in Patients with HoFH COBALT-1: Open-Label, Dose-Escalation Trial N=8 Gemcabene 300 mg Gemcabene 600 mg Gemcabene 900 mg Wk 0 -- - - -Wk 4 -- - - --Wk 8 - - - Wk 12 COBALT-1 Trial 12-week multicenter, open-label, dose-escalation trial with clinically diagnosed HoFH patients on stable low-moderate-high statin therapy (and ezetimibe and Repatha) LDL-C primary endpoint Up to 8 patients at 300 mg, then 600 mg, then 900 mg every 4 weeks 5 patients on treatment with additional patients screening at 9 sites in US, Canada, and Israel Interim data provided January 30, 2017, and top-line data results in June 2017 Regulatory pathway precedent set by Juxtapid, Kynamro, Repatha, statins, and ezetimibe for approval in HoFH on LDL-C endpoint FDA Orphan Designation granted (in 2014) Interim COBALT-1 data: gemcabene 600 mg lowered mean LDL-C by 28% on top of maximum statin lipid-lowering therapy in 2 HoFH patients Gemcabene’s novel MOA is complementary to the LDL receptor MOA’s Preclinical data demonstrated 55% LDL-C reduction as gemcabene monotherapy (and additive to atorvastatin 72% LDL-C) in HoFH mice model Clinical data (Trial 1027-018) shows up to 31% LDL-C lowering on top of stable statin therapy Gemcabene efficacy observed in difficult-to-treat HoFH patients may support potential efficacy in ROYAL-1 COBALT-1 Phase 2b Interim Results HoFH Patients on Maximally Tolerated Lipid Lowering Therapies

9 ROYAL-1 Trial Design (GEM-301) Phase 2b Clinical Trial in Patients with ASCVD/HeFH ROYAL-1 Trial 12-week multicenter, double-blind, placebo-controlled trial in patients with hypercholesterolemia (ASCVD/HeFH) on stable moderate- and high-intensity statin therapy (with or without ezetimibe) LDL-C primary endpoint at 12 weeks 104 patients (52 in each arm with target commercial dose gemcabene 600 mg and placebo) Each arm balanced 26 moderate intensity statin and 26 high intensity statin 28 sites in US Enrollment ahead of schedule (~2 months, expected completion in January) Top-line data now expected in 3Q 2017 Regulatory pathway precedent set by PCSK9s for approval with LDL-C endpoint for these high risk patients on maximally tolerated statins assuming favorable benefit/risk

Phase 2 Clinical Data Supports Potential for Meaningful Efficacy Rationale for Potential Meaningful Results in ROYAL-1 10 Phase 2 Trial 1027-018 Randomized, double-blind, placebo-controlled, multicenter trial in hypercholesterolemic subjects to determine efficacy and safety of gemcabene as add-on to stable statin therapy for patients not at goal All doses of statins were utilized (77% of patients were on moderate to high intensity statins) Primary endpoint was met for significant LDL-C reduction, up to 31% lowering Secondary endpoints were also met, including lowering of ApoB, TG, VLDL-C and TC as well as hsCRP Recently published in the Journal of Clinical Lipidology (Stein et al, 2016)

Gemcabene Shows No DDI Effects When Combined with High Dose Statins Phase 1 Trials A4141002 and 1027-008 Gemcabene in combination with atorvastatin or simvastatin was observed to be well tolerated in healthy volunteers Exposure was similar for 80 mg of atorvastatin alone or in combination with either 300 or 900 mg of gemcabene Exposure was similar for 80 mg of simvastatin (a CYP3A4 substrate) alone or in combination with 900 mg gemcabene Over 150 patients have received a high intensity statin therapy and gemcabene Rationale for Potential Meaningful Results in ROYAL-1 (cont.) 11

Rationale for Potential Meaningful Results in INDIGO-1 12 INDIGO-1 Trial Design (GEM-401) Phase 2b Clinical Trial in Patients with SHTG INDIGO-1 Trial 12-week multicenter, double-blind, placebo-controlled trial with patients with SHTG (TG > 500 mg/dL) with or without background statin therapy TG primary endpoint at 12 weeks 90 patients (30 patients per arm of 300 mg (target commercial dose), 600 mg, and placebo) 30+ sites in US and 3 sites in Canada Top-line data expected in 4Q 2017 Regulatory pathway precedent set by fibrates, fish oils, niacin, and statins for approval for reduction in risk of pancreatitis on a TG surrogate endpoint Phase 2 Trial 1027-004 Phase 2 clinical data supports potential for meaningful efficacy (Trial 1027-004) with ~40% lowering at 300 mg dose Post-hoc analysis in limited subset with baseline TGs > 500 mg/dL with reductions of ~60% at 300 mg dose Observed to be well-tolerated in combination with statins, unlike fibrates Phase 2 Trial 1027-004 Results TG > 200, GEM Lowers TG 39%; TG > 500, GEM Lowers TG 60% INDIGO-1 Double-Blind, Treatment Phase Gemcabene 300 mg N=30 Placebo N=30 Day 1 - - - - - - - - - - - - - - - Wk 12 RANDOMIZATION Screening Phase TG > 500 mg/dl Gemcabene 600 mg N=30 Note: Although patients treated with gemcabene at 600 mg and 900 mg were observed to have lower triglycerides, the lowering effect was not significant when compared to placebo.

13 AZURE-1 (GEM-501) Phase 2 Trial Focused on NASH Preclinical and Clinical Data Support Further Pursuit of NASH Indication AZURE-1 Plan* Phase 2 trial design being finalized with plans to initiate AZURE-1 in 2017 Up to 24-week multicenter trial in patients with NASH Diagnosed by biopsy within 12/18 months, Fibrosis stage <3 Steatosis by MRI-PDFF>10% Estimated 81 patients (27 patients per arm of 300 mg, 600 mg, and placebo), with interest from COBALT / ROYAL / INDIGO sites Primary Endpoint will be percent change in hepatic fat fraction determined by MRI-PDFF evaluation (see chart) Regulatory pathway evolving with FDA working with sponsors but currently Phase 3 endpoint focusing on NAS and/or fibrosis stage changes is acceptable Rationale for Potential Meaningful Results in AZURE-1 Preclinical data demonstrated reduction in lipogenesis and hepatic fat content Preclinical data in STAM model demonstrated reductions: liver NAFLD activity score (NAS) (a composite measure of fatty liver disease comprised of measures of steatosis, inflammation, and hepatocyte ballooning), progression of fibrosis, markers of inflammation, and lipid modulation Clinical data from Phase 2 trials showed reductions on lipid parameters (e.g. TG, LDL, apoB) and on inflammation (e.g., hsCRP) Gemcabene lowered TG (and cholesterol) synthesis pathway, likely inhibiting ACC as a fatty acid mimetic, resulting in reduced TG in liver Gemcabene lowered plasma TG (via APOC-III) in patients with elevated TGs; reduced APOCIII enhances remnant clearance and LPL activity Gemcabene has not shown any liver toxicities at doses between 150-900 mg up to 12 weeks as monotherapy or combined with statins/other drugs across 895 patients; no cumulative toxicities in 26 to 52 week monkey studies *Current design pending final discussions with Health Authorities

Lowering LDL-C Decreases CV Risk Elevated LDLC is the #1 Modifiable Risk Factor Sources: CTT Cholesterol Treatment Trialists and Study Papers for each Trial MACE = Major Adverse Cardiovascular Events * A-Z p=.14 and IDEAL p=.07 Decades of Evidence for the LDLC Hypothesis Grow Stronger LDLC Surrogate Endpoint Precedent for Traditional Approval of Lipid Drugs 14 TG hsCRP LDLC Lowering Drugs with Successful Trials: Gemfibrozil: HHS; Atorvastatin: IDEAL, TNT, PROVE-IT, ASCOT-LLA, SPARCL; Rosuvastatin: JUPITER; Simvastatin: A-Z, HPS, 4S; Pravastatin: ALLHAT, CARE, PROSPER, LIPID, WOSCOPS; Lovastatin: AFCAPS; Ezetimibe: IMPROVE-IT Over past two decades, all statins and other lipid-lowering drugs, including ezetimibe, were approved on the LDLC endpoint with broad labels without CV outcomes trial (CVOT) in US and globally In US the bar was raised in summer 2015, Praluent approved for HeFH/ASCVD and Repatha approved for HeFH/ASCVD/HoFH on the LDLC endpoint on maximal tolerated statins; NOT approved for monotherapy (statin-intolerant) and/or primary patients In contrast broad labels for Repatha and Praluent were approved on LDLC endpoint, including monotherapy (statin-intolerant) and primary patients (familial and mixed dyslipidemia) in Europe Repatha’s FOURIER CVOT met the primary endpoint for CV risk reduction, continues to confirm and strengthen the LDLC endpoint for regulatory approval in high risk patients

Gemcabene Novel MOA and Clinical Safety 15

Clearance Mechanism: Gemcabene Molecule Plasma half life of 32 to 41 hours Liver is target organ Gemcabene is the active compound Renal elimination *Potential molecular targets in the liver (ApoC-III, ACC) Production Mechanism: Gemcabene Novel Mechanism of Action 16 Triglyceride Pathway Cholesterol Pathway ApoC–III VLDL LDL LDL Receptors Acetyl-CoA HMG-CoA Malonyl-CoA Fatty-acyl-CoA Cholesterol Triglycerides VLDL Remnant Acetyl-CoA carboxylase (ACC)* Mevalonate Remnant Receptors (ApoE) Acetoacetyl-CoA Liver VLDL ~ ApoC–III* ~ LDL TG ~ ApoE ~ ApoE ~ ~ Plasma Gemcabene reduces production of cholesterol and triglycerides pathways inside the liver Gemcabene clears VLDL efficiently due to a reduction in ApoC-III followed by rapid uptake by the remnant receptor HO O OH O O

Gemcabene Inhibits de novo Synthesis of Both Cholesterol and Triglycerides Gemcabene has been shown to reduce hepatic de novo cholesterol and TG synthesis from acetate MOA Evidence: Production Mechanisms Reductions in TG (Fat) Seen Across Models 17 Gemcabene Reduces Hepatic Triglycerides in Sprague Dawley Rat Model (from [14C]-Acetate in Primary Rat Hepatocytes) Gemfibrozil reduces TG and other NASH biomarkers in human trials; potentially promising for gemcabene Gemcabene significantly reduces hepatic triglycerides by -74% in a rat model similar to reductions seen by gemfibrozil In-vitro Rat Liver POC TG In-vivo Rat Liver POC TG In-vivo Human POC TG A Controlled Trial of Gemfibrozil in the Treatment of Patients with NASH -74% TG’s Source: Heart Study 16, Parke Davis P<0.0092 P<0.0094 P<0.0067

Gemcabene (300mg/day) (Trial 1027-004, 12 weeks) Gemcabene Reduces Plasma ApoC-III and TG in Humans MOA Evidence: Clearance Mechanisms Gemcabene Potential Targets ApoC-III, not PPAR alpha 18 ApoC-III protein is causal for cardiovascular disease Lowering ApoC-III enhances VLDL clearance and reduces LDL-C Gemcabene potentially binds to NFkB, a transcription factor, and disrupt its interaction with the promoter region common for both the ApoC-III and hsCRP genes (AHA 2015 poster) Recent in vitro assays for the FDA show gemcabene has little to no direct binding to the mouse, rat, and human PPAR alpha (human shown in chart), PPAR gamma, and PPAR delta when compared to known reference agents; PPAR effects seen in rodents are likely secondary Nevertheless, given our classification by the FDA as a PPAR agonist, which limits our clinical exposure up to 6 months until studies are completed, we have initiated 2 year mice and rat carcinogenicity studies In-Vitro PPAR Receptor Assay (Human) - 14% - 31% - 49% - 15% - 39% - 60% -70% -60% -50% -40% -30% -20% -10% 0% TG<200mg/dL (n=11) TG>200mg/dL (n=21) TG>500mg/dL (n=6) Percent Reduction ApoC-III TG

19 Potential Pleiotropic Mechanisms of Actions Lipid Metabolism, Inflammation, Atherosclerosis, Glucose and NASH/Fibrosis Glucose Metabolism Insulin Sensitivity Atherosclerosis IL-6 & IL-1 NASH Inflammation Ballooning Fibrosis NAS Score Lipid Metabolism LDL-C TG’s (ApoC-III) VLDL HDL-C ACC Inflammation HsCRP NFB CCR2 & CCR5 Gemcabene

Gemcabene Was Observed to be Well Tolerated in Clinical Trials Adverse Events (AE) Rates Comparable Across All Treatment Groups Gemcabene was well tolerated at single doses up to 1500 mg and multiple doses up to 900 mg/day for up to 12 weeks (837 subjects) Safety evaluation of AE monitoring, clinical lab assessments, ECGs, physical exams and vital sign assessments were conducted across all trials (1,289 adult subjects): 10 healthy volunteers or patients reported a treatment-emergent serious adverse effect (SAE), none of which were related to gemcabene No deaths occurred AEs were generally mild to moderate (e.g., headache, weakness, nausea) No myalgia (muscle effects) as monotherapy, no increase in myalgia when added to statin Small mean increases in serum creatinine and blood urea nitrogen (BUN) observed in some trials, reversible within approximately two weeks of cessation of gemcabene No clinically meaningful changes in liver enzymes (0.23% of gemcabene patients compared to 0.26% of placebo patients had ALT or AST > 3 times upper limit of normal) No clinically meaningful changes were observed in physical examinations, blood pressure, vital signs and ECGs 20

NASH Proof of Concept in STAMTM Model 21

NASH Preclinical Mouse Model Study Design STAMTM: In Vivo Predictive Pharmacology Model of NASH and HCC 22 Fibrosis HCC NASH Steatosis 100% 100% 100% G-1 G-2 G-3 G-4 G-5 G-6 Sacrifice Total = 48 mice 4 wks Oral, QD for 3 weeks Oral, QD for 3 weeks Oral, QD for 3 weeks Oral, QD for 3 weeks Oral, QD for 3 weeks Oral, QD for 3 weeks (n=8) (n=8) (n=8) (n=8) (n=8) (n=8) 5 wks 6 wks 7 wks 8 wks 9 wks Normal + Vehicle STAM + Vehicle STAM + Gemcabene STAM + Gemcabene STAM + Gemcabene STAM + Telmisartan 30 mpk 100 mpk 300 mpk

23 Gemcabene CVC – Tobira** NAFLD Activity Score (NAS) and Fibrosis Score Gemcabene Demonstrates Significant Efficacy in STAMTM Model NAS* score decreased 1.2 to 2.3 pts (25% to 48%) compared to vehicle NAS* score decreased 1.3 to 1.6 pts (25% to 30%) compared to vehicle Liver fibrosis reduced ~33% compared to vehicle Liver fibrosis reduced ~60% compared to vehicle E. Lefebre et al., The Liver Meeting AASLD, Abstract 30 presentation, 2013 *NAS composited comprised of steatosis, inflammation, & ballooning **This comparison is for illustrative purposes as these were separate studies Full details available under CDA or upon publication FXR Agonists Intercept’s OCA (10 mg/kg) decreased NAS* score 1 pt (23%) compared to vehicle Enanta’s EDP-305 (3 -10mg/kg) decreased NAS* score ~1.6 pts (~30%) compared to vehicle Enanta Pharmaceuticals Company Presentation, 2016

24 Gene and Serum Biomarkers in STAMTM Model Gemcabene Demonstrates Significant Reductions in Relevant Biomarkers Hepatic Gene Expression and Plasma Markers Indicative of Inflammation (e.g., CRP and CCR2/CCR5) and Lipid Modulation (e.g., ApoC-III and ACC1) were Significantly Reduced Vehicle in NASH Gemcabene 100 mg/kg (vs Vehicle in Normal) (vs Vehicle in NASH) CRP - Inflammation CCR2 Inflammation CCR5 Inflammation ApoC-III Lipid Metabolism ACC1 - Lipid Metabolism Plasma TG and hsCRP were Significantly Lowered in NASH Preclinical Model, Consistent with Gemcabene Effects Observed in Clinical Trials Full details available under CDA or upon publication - no significant difference ? significant increase ? significant decrease

Competitive Differentiation, Commercial Strategy, Team, and Milestones 25

Limitations of Current Standard of Care in Dyslipidemia 26 Gemphire is well positioned with its product profile to capitalize on the large unmet market opportunity in dyslipidemic patients Patients Physicians Payors Lack of Convenient (injectable PCSK9s) and Tolerable (muscle, liver, GI, neuro, etc.) Therapies Limited Safe and Effective Combinable Drugs with Statins for Lipid Management (LDL-C, TG) Lack of Cost-Effective Options with PCSK9s and Orphan Priced Drugs (Juxtapid, Kynamro)

27 Gemcabene 2nd Line Oral LDL-C Positioning Differentiated Profile for ASCVD/HeFH and HoFH Patients Statins (Lipitor, Crestor, Zocor) Doubling statin dose only adds 4%-6% efficacy with increased toxicities Ezetimibe (Zetia) Juxtapid (HoFH) Gemcabene ETC-1002 CETPi Repatha Praluent Kynamro (HoFH) ALN-PCSsc 1st Line Oral Standard of Care 2nd Line Oral Approved Drugs (and Product Candidates in Development) 3rd Line Injectables (PCSK9s and Antisense) Gemcabene is a differentiated drug candidate as an add-on to all doses of stable statin therapy that offers lowering of LDL-C, inflammation, and triglycerides particularly for ‘diabesity’ patients

Gemphire Differentiated in Both Dyslipidemia & NASH Dyslipidemia Gemcabene Phase 2b NASH Statin Class Approved PCSK9 Antibodies Approved MDCO PCSK9si P3 ready ESPR ETC-1002 P3 started Fibrates Approved Fish Oil Approved Ionis Volanesorsen P3 (FCS, FPL) ICPT OCA P3 (Approved PBC) Genfit Elafibranor P3 Gilead GS-4997 P3 ready Nimbus / Gilead GS-0976 P2 Tobira CVC P3 ready Conatus Emricasan P2 MOA HMG-CoA reductase inhibitor PCSK9 inhibitors RNAi ACL inhibitor PPAR-agonist Omega-3 APOC-III antisense inhibitor ApoC-III, also ACC inhibitor FXR agonist PPAR- & agonist ASK-1 inhibitor ACC Inhibitor CCR2/CCR5 inhibitor Caspase Inhibitor MOA Once-Daily Oral  r r   O r O Once-Daily Oral Low Cost  r r  r NAS Score Reduction Preclinical LDL  r r r  unknown Anti-Fibrotic Effect TG r r r  r r Fat / TG hsCRP  r r r O r  unknown r  unknown  hsCRP Combine Safely w/ Statins  r r Unknown O unknown unknown unknown r unknown Combine Safely w/ Statins Sources: Gemphire Estimates, ClinicalTrials.gov, Analyst Research Reports, Company Websites and Presentations 28 Yes r No O Somewhat Not Applicable

Recent Events for PCSK9i Drug Class Competitor Scarcity and Payor Market Dynamics Amen Announces Repatha (Evolocumab) Significantly Reduced The Risk Of Cardiovascular Events In FOURIER Outcomes Study PR Newswire February 2, 2017 Courts Grants Permanent Injunction For Infringement Of Amgen’s Reaptha Patents Original Investigation August 16, 2016 Cost-effectiveness of PCSk9 Inhibitor Therapy in Patients with Heterozygous Familial Hypercholesterolemia or Atherosclerotic Cardiovascular Disease Dharuv S. Kazi, MD, MSc, MS1.2.3.4.5; Andrew E, Moran, MD, MPH Pamela G. Coxsonm, PhD 123; Joanne Penko, MS, MPH12; Daniel A. Ollendorf, PhD; Steven D. Pearson, MD, MSc, Jeffery A Tice, MD2; David Guzman, MSPH1; Kirsten Bibbins-Domingo, PhD, MD, MAS12345 Medscape Medical News Neurology Pfizer Stops Development of Novel PCSK9 Inhibitor, Halts Ongoing Trials Deborah Brauser November 01, 2016 August 2, 2016 10:54 AM EDT Pfizer Backs Off Competitive PCSH9  29 Amgen’s Repatha met the primary endpoint of CV risk reduction in the FOURIER CVOT, as well as the secondary endpoint of CV death, non-fatal MI, or non-fatal stroke Cost-effectiveness in HeFH/ASCVD determined at ~$4500/year by JAMA August paper Limited sales after launch for Repatha and Praluent due to access and price (2016 analyst estimates PCSK9i combined sales ~$200-300M) Pfizer discontinued injectable PCSK9 due to high level of immunogenicity (not fully humanized) and attenuated LDL-C lowering Praluent (by Sanofi/Regeneron with sales estimates of $1-3B) could be pulled from the market if appeal is denied, leaving a single PCSK9i drug Repatha with no pricing competition (~$14K current list price) Pfizer discontinued oral PCSK9 program

Recent Dyslipidemia and NASH M&A Deals by Pharma Gemcabene Positioned Across Both Markets Allergan to buy Tobira in deal valued at up to $1.7 billion published: Sept 20, 2016 8:10 a.m. ET September 21, 2016 Allergan Snaps Up Akarna in Second NASH-Related Purchase This Week UPDATED: Gilead bags early-stage NASH drug in $12.B Nimbus deal by John Carroll | Apr 4, 2016 8:46am CASH For NASH Caspase embraced: Novartis, Conatus $700M deal proves once-doubted class is no FXR-upper By Randy Osborne Novartis in $1.6 bln deal for Ionis, Akcea drugs by Denise Roland Published: Jan 6, 2017 6:26 a.m. ET September 16, 2015 For Up to $1.55B, Amgen Acquires Dezima Pharma 30 Total Deal up to $700M with tiered double-digit royalties Up front $50M Total Deal up to $1.6B Up front/Equity $175M Total Deal up to $1.55B Up front $300M Total Deal up to $1.7B Up front $600M Total Deal up to $1.2B Up front $400M Dyslipidemia NASH Total Deal $50M+ with milestones/ royalties undisclosed Up front $50M Phase 2 Preclinical Phase 1 Phase 2 Phase 1/2 Phase 2

Maximize the Global Commercial Value of Gemcabene Discussions with Regional and Global Strategic Partners Ongoing 31 Commercial In US, we may commercialize gemcabene for the orphan indication of HoFH with our own targeted sales force to 50 lipid centers and 500 lipidologists In US, we may directly sell or co-promote with a partner gemcabene for SHTG with our internal sales force and distributor(s) In US, we may partner to commercialize gemcabene in the larger indications such as HeFH and ASCVD Outside of the US, we would plan to seek global and regional partners to commercialize in key markets for all indications Gemphire has retained all global commercial and manufacturing rights to gemcabene Strategic Partnering Clinical Development We may consider co-development of gemcabene in Phase 3 (CVOT for example) We may consider co-development of gemcabene in Phase 2/3 for NASH

Mina Sooch, MBA Chief Executive Officer Charles Bisgaier, PhD Chief Scientific Officer & Cofounder Jeff Mathiesen, CPA Chief Financial Officer Lee Golden, MD Chief Medical Officer Seth Reno, MBA Chief Commercial Officer Daniela Oniciu, PhD VP, Manufacturing & Preclinical R&D Rebecca Bakker-Arkema, RPh, MS, FAHA VP, Drug & Clinical Development Liz Masson VP, Clinical Operations Prior Marketed Products Experience Prior Pipeline Development Experience CER-001 (Cerenis) CER-209 (Cerenis) ETC-1002 (Esperion) ETC-216 (Esperion) PNT-2258 (ProNAi) ACP-501 (AstraZeneca/AlphaCore) Proven and Successful Management Team Many Worked Together at First Esperion and Pfizer 32

John Kastelein, MD, PhD Amsterdam, Netherlands Evan Stein, MD, PhD Illinois, USA Rob Hegele, MD Toronto, Canada Dirk Blom, PhD Cape Town, South Africa Harold Bays, MD Kentucky, USA Peter Toth, MD Illinois, USA Jay Horton, MD Texas, USA David Cohen, MD New York, USA Rohit Loomba, MD California, USA Brian Krause, PhD Michigan, USA Gerald Watts, PhD Perth, Australia Todd Leff, PhD Michigan, USA Kevin Williams, MD Pennsylvania, USA Key Opinion Leaders Involved in CVD & NASH Drug Development Distinguished Gemphire Advisory Board 33 Dyslipidemia NASH Mechanism

Report interim data from COBALT-1 Phase 2b trial Repatha reported positive FOURIER CVOT Report top-line COBALT-1 Phase 2b trial results Submit NASH preclinical and other clinical abstracts and manuscripts for publication 1H 2017 Hold ROYAL, INDIGO, and COBALT EOP2 meetings with FDA Launch Phase 3 programs in Dyslipidemia 1H 2018 Report top-line ROYAL-1 Phase 2b trial results Report top-line INDIGO-1 Phase 2b trial results Initiate AZURE-1 Phase 2 trial in NAFLD/NASH Presentation(s) at industry meetings (if accepted) Complete in-life 2 year rodent carcinogenicity studies 2H 2017 Key Catalysts Through First Half of 2018 34 2017 Potential Transformational Year with Data Readouts in All 3 Dyslipidemia Trials

Appendix 35

No upfront payment 15% equity grant at first round of equity financing Future payments totaling up to $37 million upon completion of various milestones including regulatory approvals and key sales levels Future milestone payments under the Pfizer Agreement, if any, are not expected to begin for at least several years and extend over a number of subsequent years. Tiered royalties on country by country basis based upon annual amount of net sales Pfizer License Agreement License Provides Worldwide Exclusive Rights and License to Gemcabene 36 Completed multiple MOA studies with PPAR and lipid metabolic pathways Completed multiple exploratory efficacy studies in mice, rats and monkeys Completed over 30 nonclinical GLP toxicology studies, including: 26-week repeat dose study in rats and monkeys 52-week repeat dose study in monkeys Completed 11 Phase 1 clinical trials, including: Trial 1027-003 Trial 1027-008 Trial A4141002 Completed 7 Phase 2 clinical trials, including: Trial 1027-018 Trial 1027-004 Trial A4141001 Gemcabene (CI-1027) Program

Market Exclusivity, Patents and Manufacturing 37 POTENTIAL FOR MARKET EXCLUSIVITY (New Molecular Entity) U.S. (5 years or more); U.S. HoFH Orphan (7 years); Europe (up to 10 years); Japan (up to 10 years) EXPANDING INTELLECTUAL PROPERTY ESTATE In total, 28 issued patents (4 in US) and 23 pending applications (9 in US) Original patents in-licensed from Pfizer directed to composition, formulations, and combinations Gemphire filed applications since 2011 around novel methods as a result of mining clinical data from trials 1027-004, 1027-018 and A4141001 (respectively below) SHTG: Method for Treating Pancreatitis – US Patent #8,846,761 (expiry 2032) Add on Stable Statin Therapy: Methods for Reducing CV Risk – US Application #14/370,722 (filed 2013) Treatment of Mixed Dyslipidemia and NASH – US Provisional Application #62/252,195 (filed 2015), PCT filed 2016 Gemphire filed additional applications on FDC formulations and improved manufacturing process Fixed Dose Combination Formulations – US Provisional Application #62/252,147 (filed 2015), PCT filed 2016 Processes & Intermediates for Manufacturing – US Application #14/942,765 (filed 2015) _____________________________________________________________ GEMCABENE MANUFACTURING Drug substance and drug product manufactured to GMP specifications Cost-effective manufacturing, drug substance scalable to 100 kg to meet commercial needs

Note (*): Addressable market for rest of the world is estimated by extrapolating from the U.S. addressable market. Definitions: M = millions, CV = cardiovascular, TG = triglycerides. Source: Company estimates. US ~ 300 – 2,000 RoW ~ 6,000 – 45,000 Usually caused by a mutation in both alleles of the LDL receptor gene Increased risk for CV disease HoFH (Homozygous Familial Hypercholesterolemia) US ~ 0.5 – 1.5M RoW ~ 15 – 30M Usually caused by a mutation in one allele of the LDL receptor gene Increased risk for CV disease HeFH (Heterozygous Familial Hypercholesterolemia) US ~ 3 – 3.5M RoW* ~ 60 – 75M Caused by an inherited disorder, obesity, poorly controlled diabetes, hypothyroidism, etc. Increased risk for pancreatitis and other co-morbidities SHTG (Severe Hypertriglyceridemia) US ~ 5 – 6M RoW* ~ 100 – 120M Patients who have experienced or are at risk of a cardiovascular event and cannot achieve LDL-C goal Increased risk for CV disease US ~ 4 – 5M RoW* ~ 80 – 100M Patients who have experienced or are at risk of a cardiovascular event and cannot achieve LDL-C and triglyceride goals Increased risk for CV disease ASCVD (Atherosclerotic Cardiovascular Disease) NonFamilial Hypercholesterolemia Mixed Dyslipidemia LDL-C > 130 mg/dL LDL-C > 130 mg/dL 150 < TG < 500 mg/dL LDL-C > 190 mg/dL LDL-C > 500 mg/dL TG > 500 mg/dL Large Unmet Need to Help Dyslipidemia Patients Reach Goals Potentially 14M or More Addressable Patients in the U.S. – Most on Statins 38