Corporate Presentation (NASDAQ: DCTH)

Forward-looking Statements The Private Securities Litigation Reform Act

of 1995 provides a safe harbor for forward-looking statements made by the Company or on its behalf. This news release contains forward-looking statements, which are subject to certain risks and uncertainties that can cause actual results to differ

materially from those described. Factors that may cause such differences include, but are not limited to, uncertainties relating to: actions by the FDA relating to the application; the ability of the Company to respond to FDA queries related to the

application; the Company's successful inspections by the FDA or foreign regulatory agencies; the timing and results of the Company's clinical trials, our determination whether to continue a clinical trial program or to focus on other

alternative indications, and the impact of the COVID-19 pandemic on the completion of our clinical trials; the impact of the presentations at major medical conferences and future clinical results consistent with the data presented; the

Company's ability to successfully commercialize the HEPZATO and the potential of the HEPZATO KIT/CHEMOSAT system as a treatment for patients with primary and metastatic disease in the liver; our ability to obtain reimbursement for

commercialized product in various markets; the Company's ability to successfully enter into strategic partnership and distribution arrangements and the timing and revenue, if any, of the same; uncertainties relating to the timing and results

of research and development projects; and uncertainties regarding the Company's ability to obtain financial and other resources for any research, development, clinical trials and commercialization activities. These factors, and others, are

discussed from time to time in our filings with the Securities and Exchange Commission. You should not place undue reliance on these forward-looking statements, which speak only as of the date they are made. We undertake no obligation to publicly

update or revise these forward-looking statements to reflect events or circumstances after the date they are made. 1

Executive Summary Delcath aims to be the leader in targeted, safe and

highly-effective minimally-invasive treatments for patients with cancers of the liver. UNMET NEED HEPATIC DELIVERY COMPANY & LARGE MARKET CLINICAL PROGRAM LIVER CANCER SYSTEM (HDS) OPPORTUNITY Incidence US/EU HDS + Melphalan enables FOCUS

pivotal trial Near-term (mOM) Metastatic Ocular >200K primary and the Percutaneous Hepatic >$600 TAM in US/EU metastatic liver tumors Perfusion (PHP) Procedure Melanoma (mOM) Unsurpassed 1 year 1-14

Primary endpoint met per year Delivers high dose survival data chemotherapy to the NDA resubmitted Current local/regional entire liver Longer Term (CRC, ICC, Real World Evidence treatments Limits systemic exposure Pancreatic,

etc.) Cannot treat the whole Minimally invasive, >1k commercial >>$1B TAM treatments in EU liver repeatable and well- Investigator interest in Targeted to visible and tolerated

Multiple single center more than 10 other accessible tumors publications tumor types Limited in their ability to US: HEPZATO KIT retreat (Melphalan/HDS) EU: CHEMOSAT (HDS) PDUFA Date: 8/14/23 2

Liver-Dominant Cancers High incidence with poor prognosis US Incidence

Incid Incid US Incidence ence ence of of of Liv Liv of Liv er er Liv er Dom Dom er Dom Dom inan inan inan inan t t Cancers Cancers t t Cancers Cancers* (p (p(p a arti rti a(p rti a alla a s s rti l e e sa t t els s t s h h s e o o h t w w o sw n n

h) )n o) wn) Up to Many patients with liver metastases are 50,000 DELCATH OPPORTUNITY+ not amenable to surgical resection 80% 15 largely due to extensive tumor burden 40,000 30,000 Liver: Common Site of Metastases 20,000 Limited Effective Systemic

Treatments Systemic therapies - low efficacy Immuno-oncology agents - become less effective in 10,000 the presence of metastases 0 Limited Overall Survival - mOM ICC mBC mNET mPC mCRC HCC Unresectable Liver Cancer Often

the life-limiting organ 1,2 3,4 7-10 *Metastatic Ocular Melanoma (mOM) , Cholangiocarcinoma (ICC) , Liver-dominant Breast Cancer (mBC) , Metastatic Neuroendocrine 3 6,7 7,13 11,12 15 Tumors (mNET) Metastatic Pancreatic Cancer (mPC) , Metastatic

Colorectal Cancer (mCRC) , Hepatocellular carcinoma (HCC) U U.S. .S. In Inc cid ide en nc ce e

Limitations of Current Liver-Directed Therapies 17 16 Trans Arterial

Chemo Embolization (TACE) Y90 Beads obstruct blood flow to tumor and elute chemo Radioactive beads delivered into the tumor 50-60k treatments per year in US (and growing) 10-15k treatments per year in US (and growing)

Effective, but tumors Diffuse disease: Many tumors are not recur & retreatment cannot be treated imageable - limited due to with a tumor-by- micro-metastases damaged vasculature tumor modality are common 4 Majority of Treatment

HEPZATO Kit: Enables Percutaneous Hepatic Perfusion (PHP)

Repeatable, safe & effective liver-focused disease control ISOLATION Hepatic venous flow is isolated, Blood Return Catheter enabling 12x increased dose Chemo Chemo Isolation Filtration (Balloon) Catheter SATURATION Liver Melphalan (chemo) treats

micro and macro lesions simultaneously Chemo Delivery Catheter Veno-veno Bypass Pump FILTRATION Proprietary filters remove greater 33 than 85% of chemo from the body 5

mOM: Beachhead Market Opportunity High Unmet Need, Favorable

Reimbursement Environment Unmet Need Low Risk Opportunity >5,000 cases of primary ocular melanoma per year FOCUS pivotal trial has met primary endpoints to 12,34 2,35 37 in the US/EU , ~50% metastasize to the liver support approval

in mOM US TAM ~800 patients, Europe ~1,200 patients Significantly improved safety profile over Gen 1 36 filter technology Median survival up to 12 months. Real world safety and efficacy demonstrated in EU 55% of

patients have no approved treatment option, most patients treated with multiple lines of therapy High Barrier to Entry Favorable US Commercial Economics Orphan indication status allows for extended Favorable US reimbursement

environment for ultra exclusivity orphan outpatient MD administered drugs HEPZATO is a combination drug device Kimmtrak (approved for ~45% of mOM population) regulated by CDER - no ANDA pathway priced at an average of $800K per

patient and st ended 1 year at >$150M run rate in the US IP prevents simple 505(B)(2) follow-on 20 US treatment centers = ~80% patients 6

History of HEPZATO Kit Development 2009 Randomized Phase III Study

Completed 2013 FDA Issues a Complete Response Letter 2021 Enrollment for FOCUS Trial Completed Q3/Q4 2023 Commercial Launch HEPZATO Kit 2006 Randomized Phase III Study Begins 2012 Gen 2 Filter CHEMOSAT Granted CE Mark 2016 EU Commercialization

Clinical Trial for Patients with Hepatic-Dominant 2023 Ocular Melanoma (FOCUS) PDUFA Date 8/14/23 7

* First Phase 3 RCT Results - Primary Efficacy Endpoint Met Response

Rates Hepatic Progression Free Survival Overall Progression Free Survival (ITT population) (IRC Assessment) (INV Assessment) Proportion of patients surviving Proportion of patients surviving 1.0 3.8 1.0 PHP 5.4 mo PHP PHP BAC P- mo 0.8 Cohort Best

alternative care 0.8 Best alternative care (N=44) (N=49) Value 0.6 0.6 hOR 36.4% 2.0% <0.001 5.4 1.6 P<.0001 P<.0001 7.0 0.4 1.6 0.4 ORR 27.3% 4.1% =0.003 0.2 0.2 0.0 0.0 0 5 10 15 20 25 30 35 40 45 50 55 0 5 1 1 2 2 3 0 5 0 5 0 Months

Months Hazard Ratio (95% CI) = 0.42 Hazard Ratio (95% CI) = 0.39 Crossover design confounded overall survival analysis - most subjects in BAC arm [57.1%] crossed over to PHP arm *Mix of mOM and metastatic melanoma with >90% patients

diagnosed with mOM - NDA 201848 Clinical Study Report dated 15 August 2012. 8

st Safety Issues from 1 Phase 3 and Resulting Improvements Safety

Issues Primarily Due to Filter Improvement Hematological toxicities led to 3 patient deaths Gen 2 Filter introduced in 2013 Gen 1 Gen 2 28 33 % Improvement Adverse Event Adverse Event Hughes 2016 Karydis 2018 Gen 1 2 G3/4 G3/4 % n % n Anemia

62.9% 44 Anemia 29.4% 15 53% Neutropenia 85.7% 60 Neutropenia 31.3% 16 64% Thrombocytopenia 80.0% 56 Thrombocytopenia 31.3% 16 61% Inappropriate patient selection and Protocol amendments were put in place for procedural

issues led to 1 patient patient selection death and other AE's Training improved ~90% liver involvement causing tumor lysis syndrome FDA required these issues be addressed prior to the start of the FOCUS trial 9

Multinational, multicenter, single-arm trial Efficacy

Endpoints: FOCUS Trial Primary: Objective Response Rate (ORR) compared to meta-analysis of IO therapy Secondary: Duration of Response (DOR), Disease Control Rate (DCR), Overall nd Survival (OS), Progression Free Survival (PFS) 2

Registration Clinical Trial for Patients with 102 subjects enrolled, 91 completed treatments at 23 centers in the US and EU mOM HEPZATO Tx every 6-8 weeks up to a maximum of 6 cycles Initially a RCT against Best Alternative

Care (BAC) Subsequently modified with FDA agreement to single-arm trial 10 FDA will view the comparisons with the 32 patient BAC arm as supportive exploratory analyses 10

2020 -'23 Initial Approvals Using ORR in Single-Arm

Oncology Trials Product Indication N ORR Gavreto (pralsetinib) Metastatic RET NSCLC 114 na ve = 70%, exp. = 57% Monjuvi (tafasitamab-cxix) Relapsed or refractory large B-cell lymphoma 71 39% mutant = 69%, wild-type = Tazverik (tazemetostat)

Lymphoma positive for EXH2 mutation 95 34% Zepzelca (lurbinectedin) Metastatic SMLC 2nd Line 105 35% Tabrecta (capmatinib) mNSCLC with mutation MET exon 14 skipping 97 naive = 68%' exp. = 41% Trodelvy (sacituzumab) 3rd Line Metastatic

triple-negative BC 108 33.3% Koselugo (selumetinib) Neurofibromatosis Type 1 50 66% Ayvakit (avapritinib) mGIST with PDGFRA exon 18 mutation 43 84% Pemazyre (pemigatinib) Previously treated ICC with FGFR2 fusion 107 36% Fyarro (sirolimus) Malignant

perivascular epithelioid cell tumor 31 39% Tivdak (tisotumab vedotin-tftv) 2nd Line cervical cancer 101 24% Exkivity (mobocertinib) mNSCLC w/ EGF mutations 114 28% Jemperli (dostarlimab-gxly) MMRD recurrent or advanced solid tumors - 2nd line

209 41.6% Welireg (belzutifan) von Hippel-Lindau disease +RCC, blastomas, or NET 61 49% Truseltiq (infigratinib) 2nd Line ICC with a FGF 2 fusion 108 23% Lumakras (sotorasib) KRAS G12C mutated mNSCLC 124 36% Rybrevant (amivantamab-vmjw) mNSCLC with

EGFR exon 20 insertion mutations 81 40% Jemperli (dostarlimab-gxly) MMRD endometrial cancer, 2nd Line. 71 42.3% Libtayo (cemiplimab-rwlc) Metastatic BCC 112 meta. = 21%, adv. = 29% Tepmetko (tepotinib) mNSCLC w/ met exon 14 152 43% Lunsumio

(mosunetuzumab) Relapsed or refractory follicular lymphoma 90 80% Krasati (adagrasib) KRAS G12C-mutated NSCLC 112 38.4% Elahere (mirvetuximab soravtansine-gynx) FR positive, ovarian, fallopian tube, or primary peritoneal 106 31.7% Jaypirca

(pirtobrutinib) Relapsed/refractory mantle cell lymphoma 120 50%

Focus Trial Success Criteria - Informed By FDA Interactions Critical

Single Arm Efficacy End Points* Overall Risk Benefit Assessment "Clinically Meaningful" ORR** Significantly improved safety relative to first pivotal trial Trial powered to show an advantage over immuno- oncology

(IO) agents Positive trends in exploratory BAC comparisons Lower bound at 95% Confidence Interval needed to (ORR, DOR, DCR, PFS and OS) exceed 8.3% Best Alternative Enrolled Treated "Clinically Meaningful" DOR***

Care (BAC) Arm N=42 N=32 >6 months Dacarbazine 1 0 7 1 Ipilimumab 8 6 Pembrolizumab 26 25 TACE * Per FDA and SAP ORR is the primary endpoint and per FDA primary analysis population will be treated patient population (SAP defined ITT as

primary analysis population) ** FDA did not object to using a meta-analysis of checkpoint inhibitors "to provide support for a clinically meaningful ORR" (476 patients from 16 publications, 95% Confidence Interval for ORR of 3.6% - 8.3%)

*** FDA specified that DOR would be the critical secondary endpoint and requested that patients be followed for at least 6 months to assess durability of response 12

FOCUS Trial Analysis: Prespecified Endpoint Met* ORR Advantage Coupled

With Meaningful Duration of Response ORR and DCR in the Treated Population DOR in the Treated Population PHP (n=91) BAC (n=32) Efficacy Endpoint PHP (n=91) BAC (n=32) p Value Median DOR, months 14 NC ORR, n (%) 33 (36.3) 4 (12.5) [95% CI]

[8.31-17.74] [6.93-NC] 0.0133 [95% CI] [26.44, 47.01] [3.51, 28.99] 33 4 Patients with confirmed CR or PR (7 CR, 26 PR) (all PR) DCR, n (%) 67 (73.6) 12 (37.5) Patients with subsequent PD, n (%) 16 (48.5) 1 (25.0) 0.0005 [95% CI] [63.35, 82.31]

[21.10, 56.31] Censored, n (%) 17 (51.5) 3 (75.0) Fisher's Exact Test 26.44% >> 8.3% prespecified threshold** 14 Month Duration of Response Exploratory comparison versus BAC supportive 7 Complete Responses * 02-Dec-2022 data cut,

data continues to mature and patients will be followed at least through May, 2023 * * Meta-analysis of checkpoint inhibitors (476 patients,16 publications) calculated a 95% Confidence Interval for ORR of 3.6% - 8.3% 13

Progression Free Survival Kaplan Meier Curves in Treated Populations*

Pre-Specified Exploratory Analyses* Secondary Endpoint PHP (n=91) BAC (n=32) p Value Median PFS, months 9.03 3.12 0.0003 [95% CI] [6.34, 11.56] [2.89, 5.65] PFS status, n (%) Events 67 (73.6) 25 (78.1) Censored 24 (26.4) 7 (21.9) Hazard

Ratio Estimate 0.39 0.0002 [95% CI] [0.23, 0.63] Log-Rank test. Chi-Square test. Exploratory comparison versus BAC supportive * 02-Dec-2022 data cut, data continues to mature and patients will be followed at least through

Overall Survival Kaplan Meier Curves in Treated Populations*

Pre-Specified Exploratory Analyses* Secondary Endpoint PHP (n=91) BAC (n=32) p Value Median OS, months 20.53 14.49 0.1077 [95% CI] [16.79, 25.26] [11.10, 19.78] OS status, n (%) Events 67 (73.6) 25 (78.1) Censored 24 (26.4) 7 (21.9) Hazard

Ratio Estimate 0.68 0.1110 [95% CI] [0.42, 1.09] Log-Rank test. Chi-Square test. Exploratory comparison versus BAC supportive ** 02-Dec-2022 data cut, data continues to mature and patients will be followed at least through

Overall Survival Comparison by Best Overall Response Kaplan Meier

Curves in Treated Populations* Post Hoc Exploratory Analyses* CR PR SD PD/ND ( (N=7) (N=26) N=34) (N=24) Status of OS, N (%) Events 1 (14.3) 17 (65.4) 29 (85.3) 20 (83.3) 6 (85.7) 9 (34.6) 5 (14.7) 4 (16.7) Censored NC 28.16 19.25 11.99

Median OS (Months) 95% CI [26.71, NC] [23.46, 34.46] [15.90, 23.00] [8.18, 14.03] p-value <0.0001 CR=complete response, PR=partial response, SD=stable disease, PD=progressive disease, ND=not done, BOR=best overall response Note: NC = Not

calculable, due to the number of events within the stratum (n=1) Kaplan-Meier estimates. Log-Rank test. * 02-Dec-2022 data cut, data continues to mature and patients will be followed at least through May, 2023 Analysis Supports that

ORR is Clinically Meaningful

Adverse Events Are Predictable and Manageable Related Serious Adverse

Events Occurring in >5% of PHP Patients Category, n (%) FOCUS Trial (n=95) Thrombocytopenia* 14 (14.7%) Neutropenia 10 (10.5%) Febrile neutropenia 7 (7.4%) Leukopenia 5 (5.3%) *Thrombocytopenia includes thrombocytopenia (8, 8.4%)

and platelet count decreased (6, 6.3%) Neutropenia includes neutropenia (8, 8.4%) and neutrophil count decreased (2, 2.1%) Leukopenia includes leukopenia (4, 4.2%) and white blood cell count decreased (1, 1.1%) 17

Hematological Toxicities - Comparison with Previous Trials* Grade 3 or

higher Adverse Events, FOCUS Trial* Hughes 2016 Post-procedural (n=91) (n=70) 8 (8.8%) 44 (62.9%) Anemia Thrombocytopenia 27 (29.7%) 56 (80.0%) Neutropenia 27 (29.7%) 60 (85.7%) Hematological AE's consistent with European experience

*Data cut on 02-Dec-2022 In the FOCUS Trial, "febrile neutropenia" was included under the category of "neutropenia." In Hughes 2016, "febrile neutropenia" was considered its own category; 12 patients

(17.1%) experienced febrile neutropenia. 18

FOCUS Trial - Safety Comparison with Previous Trials* FOCUS Trial

Pooled Analysis of Category (N=95) Prior Studies (N=121) Patients who Withdrew due to an AE or 17 (17.9%) 46 (38%) SAE Patients who Received a Reduced Dose 12 (12.6%) 27 (22.3%) due to an AE or SAE 4.1 2.8 Average Number of Cycles Improvement in

tolerability led to a larger number of treatments * Data cut on 02-Dec-2022 19

FOCUS Trial Comparison to Other Published Studies* Median OS Median PFS

Clinical Study Type Treatment (months) (months) N Study/Publication (95% CI) (95% CI) AL 91 20.53 9.03 FOCUS Single-Arm Hepzato (16.79 to 25.26) (6.34 to 11.56) 13 Khoja et al 2019 systemic and liver-directed 10.2 3.3 Meta-Analysis 912 (based on 29

therapies (9.5 to 11.0) (2.9 to 3.6) articles) 14 Rantala et al 2019 systemic and liver-directed 12.84 Meta-Analysis 2494 - (based on 78 therapies (12.00 to 13.56) articles) 12.7 3.0 11 TN Piulats et al 2021 Single-Arm ipilimumab plus nivolumab 52

(7.1 to 18.3) (2 to 4.1) ipilimumab + (pembrolizumab AL Heppt et al 2019 Single-Arm 64 16.1 3.0 or nivolumab) 21.7 3.3 TN tebentafusp 252 (18.6 to 28.6) (3 to 5) 9 Nathan et al 2021 Randomized 16 2.9 TN control 126 (9.7 to 18.4) (2.8 to 3) TN =

Treatment Na ve, AL = Any Line *Studies from 2019 or later with >50 patients

Expanded Access Program (EAP) Institution City Principal Investigator

Status Moffitt Cancer Center Tampa, Florida Dr. Jonathan Zager Open and Enrolling Duke University Durham, North Carolina Dr. Georgia Beasley Open and Enrolling University of Tennessee Memphis, Tennessee Dr. Evan Glazer Open and Enrolling Contract

and budget are under review Thomas Jefferson University Philadelphia, Pennsylvania Dr. Marlana Orloff with the site Contract and budget are under review Stanford University Stanford, California Dr. Sunil Reddy with the site Contract and budget are

under review Ohio State University Columbus, Ohio Dr. Aslam Ejaz with the site Mayo Clinic Hospital Jacksonville, Florida Dr. Roxana Dronca / Dr. Yiyi Yan Contract and budget discussions initiated Interested to participate in EAP; need Emory