Greetings and welcome to the Delcath’s Business Update Call. At this time, all participants are in a listen-only mode. A question-and-answer session will follow the formal presentation. As a reminder, this conference is being recorded. I would now like to turn the conference over to your host, Mr. James Carbonara, Investor Relations. Please go ahead.

Thank you. And once again, welcome to Delcath Systems fourth quarter 2020 business update call. I’d like to begin the call by reading the Safe Harbor statement. This statement is made pursuant to the Safe Harbor for forward-looking statements described in the Private Securities Litigation Reform Act of 1995. All statements made on this call, with the exception of historical facts, may be considered forward-looking statements within the meaning of Section 27A of the Securities Act of 1933 and Section 21E of the Securities Exchange Act of 1934. Although the company believes that expectations and assumptions reflected in these forward-looking statements are reasonable, it makes no assurances that such expectations will prove to have been correct. Actual results may differ materially from those expressed or implied in forward-looking statements due to various risks and uncertainties. For a discussion of such risks and uncertainties which could cause actual results to differ from those expressed or implied in the forward-looking statements, please see risk factors detailed in the company’s annual report on Form 10-K, those contained in subsequently filed quarterly reports on Form 10-Q, as well as in other reports that the company files from time to time with the Securities and Exchange Commission. Any forward-looking statements included in this earnings call are made only as of the date of this call. We do not undertake any obligation to update or supplement any forward-looking statements to reflect subsequent knowledge, events, or circumstances. Before handing the call over to Gerard Michel, I want to remind listeners that there will be a slide presentation accompanying today’s call. Please click the webcast link in the press release, and the deck can be accessed at the Investor & Presentations section of the Delcath website following the call. Gerard, please proceed.

Thank you everyone for joining today. This morning, we released preliminary data from the FOCUS trial, a global registration clinical trial investigating the safety and efficacy of HEPZATO to treat metastatic ocular melanoma patients. Due to the importance of those data, we will focus this call on those results and not discuss in detail the financial results for the quarter, which were also released this morning in our fourth quarter earnings release. Of course, I and the team are available later to answer any questions about our financial results and anything else covered in the earnings release or our 2020 10-K to be filed later today. With me on the call today are John Purpura, Chief Operating Officer; Dr. Johnny John, our Senior Vice President of Clinical and Medical Affairs; and Dr. Jonathan Zager, Lead Investigator of the FOCUS study, Senior Member and Director of Regional Therapies at Moffitt Cancer Center, who has kindly joined us to answer questions about their experience treating patients with HEPZATO and his perspective on the preliminary results. Before handing the call over to Johnny who will review the preliminary content, I want to share my perspective on the data and put it in context. As most listeners know, the FOCUS trial is the second pivotal trial setting a Delcath percutaneous hepatic perfusion system with Melphalan to treat metastatic ocular melanoma patients. The first trial used an earlier generation device as well as a somewhat different procedure and patient selection criteria. An NDA submission based on data from that trial led to a complete response letter, or CRL. Leading up to the CRL, an Oncology Drug Advisory Committee, or ODAC panel voted against supporting approval based on an assessment of the benefits of treatment with an earlier version of the system did not outweigh the risks associated with the procedure. A significant portion of FDA’s presentation to the ODAC panel was focused on the FDA’s assessment of treatment-related risks, including the analysis of treatment-related deaths that occurred during the clinical trial. Therefore, a critical objective of the FOCUS trial was to maintain or improve the efficacies in the earlier pivotal trial, while dramatically improving the safety profile. I believe we have accomplished this. There is no standard of care for metastatic ocular melanoma and we chose to power the single arm trial to achieve a higher overall response rate versus checkpoint inhibitors, one of the few metastatic ocular melanoma treatment categories with a significant amount of peer-reviewed publications, which enabled a meta-analysis from which to estimate an overall response rate as a threshold for success. As Dr. John will present shortly, the preliminary data does yield an overall response rate which well exceeds the threshold. The single-arm trial was initially designed and conducted as a randomized controlled study with a best alternative care comparator arm before being amended to a single arm trial. While the amended trial was not powered to test superiority versus best alternative care, comparative analyses against the BAC arm were included in the revised statistical analysis. It is very exciting that based on the data available, patients in the HEPZATO arm had statistically significant improvements over BAC on a number of important endpoints, including ORR, PFS, and DCR. Since most of the patients in the background were treated with TACE, this trial provides a strong signal that HEPZATO may have advantages over both checkpoint inhibitors and TACE in the treatment of metastatic ocular melanoma patients. Of course, we all must recognize that not all of the data are available for analysis and therefore results are subject to change. Nevertheless, the results to-date are very encouraging. I recognize that the preliminary overall response rates released today is somewhat lower than the range of efficacy results reported from a variety of non-sponsor studies in Europe over the past few years. The use of an independent review committee, patient selection, and likely most importantly, scan frequency may account for most of these differences. The FOCUS study, which was initially designed as an RCT, used the common 12-week scan, used in most RCTs, which the FOCUS trial was originally designed to utilize. All of the European studies scan patients more frequently with some intervals as short as 6 weeks. Since the confirmed response per resist criteria requires two sequential scans showing a response, a 6 or 9-week schedule would likely be more likely to capture a shorter duration in response than that 12-week schedule. But we must keep in mind that we have easily cleared the pre-specified target of 21%, which is based on the checkpoint inhibitor meta-analysis and even more compelling, the evaluable data are incredibly strong when compared to the BAC arm, which is really the most pertinent comparison. We will not be sharing all pre-specified analyses since the data is not yet mature enough for each analysis to provide any meaningful information. Specifically, we will not yet be sharing overall survival or duration of response, primarily because for the former there are too many patients for which the EDC has not been updated with the date of death or confirmation of the patients still living, and for the latter, there are too many later scans of both arms which need review by the IRC, but we are very confident in the data points we are releasing today. The most significant question is whether the efficacy shown combined with the improved safety profile meaningfully improves the overall benefit-risk profile over what was seen in the earlier study. The preliminary safety results are consistent with what we have seen in the over 1,000 commercial treatments that have occurred in Europe since 2015 and have been reported in the four publications from Europe. In short, the adverse events of the study were well manageable or self-limiting, and there were no treatment-related deaths. The hematologic and hepatic toxicity percentages were significantly lower compared to studies using the early generation of the device, which utilized a different filter system. While the trial is still ongoing and final and complete data will not be available until later in the year, we believe these results strongly suggest that the FOCUS trial data will demonstrate a significantly improved benefit-risk profile that warrants forming the basis of our NDA resubmission to the FDA. I will now turn the call over to Dr. John to review the preliminary FOCUS data.

Thank you, Gerard, and thank you everyone for joining this morning. I would like to start with the first slide if we could have that operator. Are we able to get the first slide?

I have advanced it.

Yes, thank you. This first slide shows the total enrollment in the trial of 144 patients, 102 patients in the PHP arm and 42 patients in the best alternative care arm or BAC arm. Of these patients that were enrolled, 91 patients were treated in the PHP arm and 32 patients in the BAC arm. Our data cutoff for this analysis was March 12, 2021. And in order to meet the criteria to be included in this analysis, patients had to have at least two evaluable response time points received and evaluated by the independent radiological committee by March 12, 2021 unless the subject had progressive disease at the first imaging time point or had no scans required and none were expected after the first evaluable time point. With the application of this criteria, we had 79 PHP patients and 29 BAC patients in this analysis. Next slide please. For patients randomized to the BAC arm, investigators could choose from 1 of 4 treatments: dacarbazine, ipilimumab, pembrolizumab, and TACE. Investigators were required to choose the BAC treatment that they were going to give the patient prior to randomization. The number of patients that were enrolled and treated in each of these subgroups is shown here. Out of the 32 patients treated in the BAC arm, 25 received TACE, 6 received pembrolizumab, 1 received ipilimumab, and there were no patients that received dacarbazine. Next slide please. This slide provides some demographic data on the trial population; the mean age in the PHP group was 57 and in the BAC group it was 60. The median was 61 in both groups. In terms of gender, we had a fairly even split in the PHP arm, and it was 45% male and 55% female in the BAC arm. With the time since diagnosis of both liver metastases, we had a median of 5.29 months in the PHP arm and 2.53 months in the BAC arm. Next slide please. In this slide, we see the objective response rate for the population that we reviewed for this analysis, the preliminary analysis population, and on the right of the slide, we have the intent-to-treat population. So far, for those 79 patients in the PHP arm that were evaluated, our response rate was 32.9% with 26 responders, and in the BAC arm, it was 13.8% with 4 responders. If you add in the ITT population, for the PHP arm, the response rate was 29.2%, and in the BAC arm, it was 10.3%. Under that, we have given the 95% confidence interval for those values, and the p-value based on Chi-square for the treated population is 0.0493 and for the intent-to-treat population is 0.0198. At the bottom of the slide, we have given the breakdown of the best overall response seen in this analysis, and that’s divided into the complete responders, the partial response, stable disease, progressive disease, and we had one non-evaluable patient both in the PHP arm and in the BAC arm. Next slide please. In terms of disease control rate, again for the preliminary analysis population, the PHP percentage was 70.89% and for the BAC, it was 37.93%. If you add in the intent-to-treat population, it was 62.92% for the PHP arm and 28.21% for the BAC arm. Disease control p-value based on Chi-square was 0.002, and then the intent-to-treat population was 0.0003. Next slide please. In the slide, we have displayed the values for the progression-free survival analysis both for the PHP arm and for the BAC arm. So, the median PFS observed in the PHP arm in this analysis was 9.03 months, and in the BAC arm, it was 3.06 months. Below that, we have provided the confidence interval, the quartiles, and the p-value for this analysis of progression-free survival was 0.0004. There were certain events that occurred to be able to establish the PFS status of the patients at this time. And there were certain events that had to be censored, and we gave the breakdown in the table there. For the PHP arm, 50 patients had events and 29 patients were censored at this time point, and in the BAC arm, it was 22 patients that had events and 7 patients were censored. The hazard ratio estimate was 0.41. Moving on to the safety of the patient population and the trial that we have observed to-date, in the safety population of 94 patients, 38 patients or 40.4% of patients experienced treatment-emergent serious adverse events, the most commonly reported treatment-emergent serious adverse event were thrombocytopenia at 14.9% of patients, neutropenia at 10.6% of patients, and leukopenia at 4.2% of patients. These were transient in nature and well manageable. Five percent of patients experienced treatment-emergent serious cardiac adverse events. In all cases, the events resulted in no ongoing complications. There were no treatment-related deaths in the trial, and the safety profile of the trial was consistent with the safety profile of PHP treatment described in literature from the European experience. The slide here provides the system organ class of the most commonly seen serious adverse events. You can see that the bone marrow suppression, which includes thrombocytopenia, neutropenia, and leukopenia, added up to 22.3%; respiratory and thoracic disorders were 6.4%, and cardiac disorders were 5.3%. At this time, I would like to now introduce Dr. Jonathan Zager, our Lead Investigator of the FOCUS study, Senior Member and Director of Regional Therapies at Moffitt Cancer Center, and ask him a few questions on his impressions of the results being shown today. Dr. Zager, what are your thoughts on the preliminary data that we are releasing today?

So, I think the preliminary data is fantastic. Obviously, if I, in my rough calculations with each step of the way, in the preliminary analysis population, in the IPT population on response rates in progression-free survival and even toxicity, there is a tripling of the response rates progression-free survival, all in favor of the PHP group, and this beats our last Phase 3 trial as well. So, it’s very encouraging, and these patients are going to benefit from this procedure.

Thank you, Gerard. Based on your experience with the treatment and your conversations with other investigators of the trial, what is your opinion of the safety profile of the treatment with the results shown here and your own experience?

So, the results shown obviously show that there is some bone marrow suppression; however, and that’s the most common adverse event. However, it’s all manageable as an outpatient, and the quality of life for these patients who have a really bad problem after the procedure is excellent. I usually treat the patients on a Friday and they see me the following Monday after they are discharged from the hospital in less than 24 hours. By Tuesday, they are back to work. So, the quality of life and any of these adverse events are overwhelmingly treated as an outpatient and/or they just recover with observation. So, I think it’s very manageable, and the patients do extremely well in terms of managing any toxicities.

Thank you, Dr. Zager. I will now open up the call for questions from the audience. Operator?

Thank you. Your first question comes from the line of Marie Thibault with BTIG. Please proceed with your question.

Good morning and thank you for taking my questions. I appreciate the time today. Maybe one for Gerard first and then one for Dr. Zager. Gerard, we appreciate sort of the detail on the BAC arm as well as the predefined criteria. I am curious on your thoughts about how the FDA will view this comparison and whether that 21% kind of hurdle with the checkpoint inhibitors is the kind of comparator they are more likely to use or whether the BAC is what matters the most to them?

Well, in the correspondence with the FDA leading up to the amendment of the trial, we used that meta-analysis to show them that was the hurdle we thought we needed to get over. The FDA, of course, although they did not object to us amending the trial, they never said, 'Hey, you know, if you hit that you are approved,' as their standard response is it will be a benefit-risk evaluation on the totality of the data. I think the big question is whether or not the label will include superiority over BAC. Again, if this data holds up when all of it comes in or whether or not it will, but I am sure you know, in totality, the FDA is going to weigh quite heavily the fact that this looks like it has an improvement over likely TACE, because this is mostly TACE in the BAC arm as well as beating the meta-analysis for checkpoint inhibitors. So, it can only help whether or not it gets to the label or not, I would rather not guess at this point.

Okay, that makes a lot of sense. And I know that you mentioned that we should expect to see full results a little later this year; any idea on the timeline for that? I know you hesitate to give timelines in some cases here?

Yes. And again, the big, I know I sound like a broken record at times, the big hurdle for us is monitoring, and that was driven by the financial difficulties of the company when the EDC was shut down for a period in 2019. And when that got up and running again, COVID showed up, so double whammy. We are very hopeful that by the end of the year, probably sometime around the late third quarter, we’d have all the data and release it. I am not sure whether we will do a scientific meeting or release it for audience before that, but definitely by the end of the year.

Okay, that’s very helpful. Dr. Zager, if I could please ask you, want to thank you for your time today. On the safety events, you mentioned some of the bone marrow suppression. I am curious to get your thoughts on any of the sort of cardiac events that showed up during the trial, the seriousness of those as well as maybe your thoughts on how if the PHP treatment were to be approved, how you might adopt it within your own center? And thanks so much for the question.

So, I can comment on all that. I will take the second question first. We have already done about 130 PHPs between the last trial, this trial, and compassionate use in between both trials. So, adopting it at the center, we have already adopted it, and we are ready to offer it to patients. So that’s a no-brainer. I have seen firsthand over the last decade how it changes the patient’s lives for the better; we get control over this disease, as you saw, 70% plus or so of the time, and it’s extremely well-tolerated. As far as the cardiac events are concerned, my patients in those procedures did not have any serious adverse cardiac events. I know there were a few on the trial that could be easily explained and actually don’t seem to be related to the double-balloon catheter inflation of the balloons placing the patient on bypass pump, or the filters. They were two kind of variant events that could be seen in any patients, even those going under appendectomy or having catheterization for chemoembolization or whatnot. So, I think that again, as I mentioned before, the procedure is extremely well tolerated. Once you have an understanding of how to perform the procedure, which is not very hard to perform, there will be obviously more than adequate teaching for new centers that come on board; it’s going to be very well received in those centers.

Thank you.

I will now open up the call for questions from the audience. Operator?

Thank you. Your next question comes from the line of Scott Henry with ROTH Capital. Please proceed with your question.

Thank you. Good morning and congratulations on the data. Just a couple of questions. First, I believe you have data on 79 of 91 treated patients; what number would you expect that to get to in the final analysis? I see for the best available care you ended up with 29 of 32, and I assume that’s been done for a while, but how close are you expected to get to that 91?

In terms of full analysis, I mean, we will see...

In terms of full analysis, would you expect to just lose some data in the process or just trying to get a sense of how many patients could ultimately be added in the final analysis?

Well, they will all be added. I mean, there are patients that you end up losing the follow-up at some point, that’s normal, and then you center them. Johnny, maybe you can spend a little more detail?

Yes, currently, we expect to get all the remaining patients in the analysis as we go forward. With the BAC patients, as you pointed out, there are three patients that we are awaiting additional data on, and then there are 11 patients in our treatment. But at this time, we expect all those patients to be included in the analysis. We didn’t have them and they did not meet the criteria that we put forth for the preliminary analysis at this time point, but we plan to obtain that data for the upcoming analysis.

Okay, great. And I just want to make sure I understood this correctly; in the prepared remarks, there was discussion about how the response rate may be a little lower in this trial versus prior trials, and my understanding was that prior trials had more scanning thus it will be easier to have two in a row and get an ORR. Is that the correct interpretation of what the point you were making?

Yes, that is correct. So, this was started out as an RCT with an overall survival endpoint, and it’s typical to just use the traditional 12-week scan periodicity, which is done when you are treating patients. A lot of overall response trials do a shorter periodicity. It was not appropriate to change the periodicity for this trial, I understand, for when it was amended, because the team did not want to change too many things when they were asking to amend with the FDA, and they also wanted to be able to have a reasonable ability to compare to the BAC arm, which you wouldn’t have been able to do if you change the periodicity. So, it was kept the same. But yes, if you are scanning, as an example, every 6 weeks, you will be able to catch responses that are perhaps 9 weeks long versus if you are scanning every 12 weeks, you will lose those shorter duration responses.

Perfect. Thank you for that clarity. And then obviously, the data is preliminary, but I did want to ask, and you may not have this stratification, but are there any differences in the results over the timing of the trial from perhaps at the beginning of the trial towards the end of the trial or among centers? I know this is a procedure that can take some learning curve or is it pretty consistent throughout the data and you may not have that specificity?

Johnny, we don’t have the specificity; correct?

Yes, correct, Gerard. We haven’t done a deep dive into that specificity at this time point to evaluate new centers versus experienced centers, but also patients that were enrolled earlier at those centers and later on after subsequent experience. We will be doing that, but we don’t have this at this time.

Okay, great. And final question, the data was pretty consistent across everything; still waiting for overall survival. Based on prior trials, is there any reason to think overall survival wouldn’t correlate heavily with ORR and median progression-free survival, or in past trials has that endpoint correlated pretty similar to the other data points?

I think if you look across oncology trials, it’s probably a mixed bag. And Johnny and Dr. Zager probably could speak more intelligently than I can. I can’t say we don’t have any reason at this point to believe that it wouldn’t correlate. I will offer that I am not going to – I don’t want to play too many questions on the numbers; I will offer that the trends are certainly in our favor at this point, but there are just so many non-evaluable patients that it wouldn’t make a lot of sense to try to put the data at this point.

Okay, thank you for the color there, and thank you for taking the questions. Congratulations.

Yes.

Your next question comes from the line of Arlinda Lee with Canaccord. Please proceed with your question.

Hi, guys. Thanks for taking my questions. I have a couple. I think I might have missed it, Gerard, you mentioned that the relevant comparison was HEPZATO versus BAC? What was the response rate by ITT? And then for Dr. Zager, can you maybe talk a little bit more about the aberrant cardiovascular events that you talked about? And then maybe a third question is can you provide additional information on the number of treatments that the patients had? How many are ongoing and what the duration of follow-up is? And then for Dr. Zager, can you talk about how these patients were if they needed any pretreatment and how long they were in the clinic? Thank you.

Sure. So, picking up the first question, the trial was powered to beat a meta-analysis on immunooncology patients. The upper bound of that point estimate was roughly about 8% for immunotherapy. What we had to do was to beat that at a 95% confidence interval. And to beat that, a point estimate of 21% would do that, alright, not a lower bound or an upper bound, so I know some people are getting texts here about the confidence intervals. We needed to be at a point estimate of 21% to beat the upper bound of meta-analysis for immunooncology agents. So, we have sailed well over that bar. The BAC analyses were secondary or exploratory technically it might be, but they are predefined in the SAP. So, we think those are – that was not what the trial was powered to show. Now at the moment, we are beating it, which is great, and it’s obviously very, very supportive, both for I think regulatory purposes and also very much in terms of uptake when it’s published in the scientific literature in terms of update with the oncology community. Now I will pass it over to – I think you had a question about the cardiovascular events; Dr. Zager can answer those.

Yes. So again, I am going to take your second question first, which was pretreatment; none of these patients were pretreated. If they had, I can’t say none of them were pretreated. None of them were pretreated by the centers that were enrolling the patients, meaning that if you had immunotherapy prior or chemotherapy prior to enrolling on the trial, there is a certain washout period to make sure that one, the effects of that therapy weren’t going to be counted towards a trial patient win per se. So – but we did purposely pre-treat anybody if they came to me or the other centers with previous treatments that obviously were failing, because obviously, they are not going to come to me if those treatments were working. They had the appropriate washout period, which I think was 8 weeks for immunotherapy and 4 weeks for other therapies like Y90 and chemoembolization. Second, to the first question about the cardiac events, there were two cardiac events that were SAEs. One was after a drug, an anesthesia drug was given on the sixth PHP procedure on a patient who had done obviously extremely well considering she was getting her sixth PHP; she was having an ongoing response, tolerated the PHPs very well, and there is a drug that is known to cause severe hypotension. The patient had a cardiac event immediately after that drug was given upon or right before excavation. So, it was the timing was related to the administration of that drug. Obviously, that still counts as a cardiac event for the trial, but there is an explanation of why that happens. The second event was related to an introducer, which is the plastic catheter for which the double-balloon catheter is inserted through this introducer. So, there is some rigidity in the vein. This, without that, you can’t introduce guidewires or other catheters. So, you need this introducer to create the hole in the channel. The introducer, which can be off the shelf or part of the kit, had a faulty valve in the cap of the introducer, which is usually right up against the patient’s skin. During the procedure, I guess that faulty valve was allowing backflow of blood, and it’s in the patient’s groin. So, it’s hard to see; because that blood trickles down, this could happen in chemoembolization and Y90, in any interventional radiology procedure, and an IVC filter placement, there are like hundreds of procedures where you use these introducers, trauma, and so forth. The patient blood is a significant amount due to this faulty valve, and that caused some again hypotension in a cardiac event; not related to the double-balloon catheter to bypass the filters and the drug was never even given. So again, these events, that’s why I was saying before these are explainable, we can put our finger on exactly what we feel caused the events and not necessarily related to the device, meaning the double-balloon catheter, the filters or the need of the procedure, where we are giving chemotherapy and perfusing the liver. Does that make sense?

Yes, it does. Thank you very much.

Yes, so in the trial, a patient could get a maximum of 6 treatments in the PHP arm. If you took an average overall of the treatments to date, we have had 371 treatments. The average is between 4 and 5 at the current moment.

Great. Thanks.

Your next question comes from the line of Yale Jen with Laidlaw & Company. Please proceed with your question.

Good morning and congrats on the data. And my first question is that in terms of the safety, which is the most important aspects of it, how do you guys compare the current data versus the previous data reported a number of years ago, mainly in safety?

Johnny?

So, the primary adverse events that we are seeing, of course, as we have mentioned a few times here is the hematological toxicities. In the previous trial, we had leukopenia and neutropenia and thrombocytopenia around 80% level. We are seeing a decrease in those percentages in this trial. So in terms of bone marrow suppression, toxicities, we do feel that we will have a better safety profile in this trial, and that’s what the evidence has shown to date. We did have a few cardiac events as mentioned and detailed by Dr. Zager. Besides those two events that he spoke of, we have had some arrhythmias in patients, all of those results without sequelae, and the patients did not have any further need for cardiology treatment. This was transient in nature. But the other in terms of liver, enzyme elevations, electrolyte imbalances, all of them were very much in line with the experience we have seen in the European fields outside of the clinical trial. We are quite confident that the safety profile has not shown any areas of concern.

Well, Johnny, can we just simply say that there has been a dramatic improvement in the safety profile versus what was seen in the first trial?

Yes, we could say that, Gerard.

Okay, great. That’s very helpful.

I would just like to ask Dr. Zager if he would like to comment. He was on the first trial as an investigator and had the experience with the previous filter and has now treated a large number of patients on this trial. Dr. Zager, would you agree with the safety profile?

Yes, I definitely agree with what you said. I would like to add one other piece of information that we have some hopefully soon to be published data and submitted data to conference for a conference presentation looking at our specifically at the cardiac sequelae and any consequences after the procedure. What we found was that we actually found that there were no long-term or even short-term sequelae from any cardiac enzyme elevation, never correlated with EKG or echo changes on the – I think it was just over 100 patients that we treated at Moffitt Cancer Center, and that we are able to gather data on. So, this is unpublished data, but soon to be in hopefully accepted for presentation as a meeting and published soon thereafter. That includes the last Phase 3 trial, the compassionate use patients, and these Phase 3 patients, again just specifically looking at the cardiac issues of which we found none. As far as between the two trials, definitely I feel that the new filters, these filters have significantly improved the potential bone marrow complications and adverse events, especially the severity of those, and they are much shorter in duration than they were with these last filters, which weren’t as efficient in removing the melphalan.

Okay, great. That’s very helpful. And maybe two quick ones. The first one is that the – I noticed that the progression-free survival was quite significantly improved versus the last study, I think this time it’s 9 versus 3.1 months versus the last time it was 4.7 versus 1.6 months. Any comments on this particular aspect?

Johnny?

Well, yes, you are right that we have noticed this in the current trial. We had to, of course, sensor some patients. We hope to see this evolve over time. But again, it goes to the improved procedural and safety measures that have been taken place since the last trial and this trial. We have incorporated the lessons learned from the other trial in terms of the safety measures and doing the procedures and treating the patients after the procedure. So, all these are reflected in these results that we see. We also take learnings from our European experience and we do follow cases there that are treated outside of the trial and provide consultation with those investigators and doctors that are treating patients in the European market, and then incorporate those into our protocol. So, when the protocol was written and the amendments were put in place, we did incorporate those measures for these patients, and so we do see the results of that. Another measure is that so far to date we have had no treatment-related deaths that are attributed to the treatment. We hope that the overall survival that we will evaluate at the end, as we have more data comes in, will also reflect these improvements.

Okay, great. Maybe the last question for Dr. Zager, which is that you guys have done a tremendous number of procedures, so you are more familiar with the process, but what do you think about the going forward, you said procedures get approved for new centers; that centers have not experienced using this procedure, how easy or how difficult for them to adapt to the PHP process – procedures? Thanks.

Understood. So, there will be a program of proctoring and teaching new centers the procedure as they come on board just like there was with some of the expert centers at the beginning of this Phase 3 trial, we went ahead proctored and taught centers how to do the procedure. So, that program will continue. It’s just like any other new technology, new procedure; there needs to be some peer-to-peer proctoring just like if it were a new robotic type of surgery or a technique. As far as adaptation, the process of putting in the catheters and in performing the procedure and technical aspects is extremely easy for the interventional radiologists. The procedure has a couple of steps during the procedure that need to be kind of just detailed with the new team coming on board, but very easy to understand and follow, and obviously a few things along the way during the procedure that you want to – some checkpoints that you want to make sure that are done just like any other surgical or interventional procedure. So, it’s easy to teach. It’s extremely easy to perform. And I cannot imagine that it won’t be adopted very easily, very quickly in centers. I mean, this is in my opinion, it’s an extremely well-tolerated procedure that provides obviously based on today’s presentation a benefit to patients who have extremely historically poor survival with ocular melanoma metastatic to liver, and now it gives them something that’s extremely well-tolerated that gives them a chance to have some prolonged progression-free and overall survival based on our preliminary dataset.

Okay, great. Thanks a lot, and again congrats.

Thanks.

Your next question comes from the line of Swayampakula Ramakanth with H.C. Wainwright. Please proceed with your question.

Thank you. This is RK from H.C. Wainwright. Thanks, Gerard and Johnny John for doing this. I have couple of quick questions. To me it seems like most patients from at least from the numbers that I remember of the background; most patients were actually treated with TACE at least in the back, and those 7 patients treated with immune checkpoint inhibitors. So, how if you haven’t had a chance to see what would it be if you just compare it only against TACE?

Yes, we haven’t done that, and the numbers start getting rather thin obviously, but I think it’s just important to say. I think Dr. Zager probably thinks TACE probably works better than the checkpoint inhibitors. And that might be why you saw the BAC control arm do a bit better than in the previous trial, but we haven’t sliced it; but I think it is important. And there is not much in the way of data in TACE published at all. So, this is probably one of the first trials that actually tested TACE to get something else. But again, we are very encouraged that it looks like, again, its part of the group of best alternative care, but we are encouraged that there seemed to be at least a signal you could say that we are doing better than TACE. I don’t know Dr. Zager, maybe you can speak to what’s in the literature about TACE and what do you think about this versus TACE to the extent you can even comment at this point?

Yes. So, the last trial, best alternative care since it was a crossover, the best alternative care was a mixed bag of possible treatments, so it’s hard to compare just to TACE. This trial obviously I think 70% something of the best alternative care patients were in the TACE group. So, the numbers are representative of maybe what TACE can do for the patients, again, inferior to PHP. As far as a direct comparison, there is one paper out from my institution that directly compares PHP versus chemoembolization and TACE. The numbers were in favor hepatic progression-free, progression-free and overall survival, all in favor of PHP, but with the retrospective outcomes based comparison of these patients were very small numbers. You can look up the paper and you can see that the numbers were definitely in favor that PHP gives the patient a significantly improved hepatic overall progression-free and overall survival versus the other interventional radiology options for these patients. But other than that, the literature is not very robust in terms of looking at TACE specifically for ocular melanoma patients. There is some smattering of reports here and there, but I still think PHP outperformed it.

Thanks for all that color. And what was the time starting from when the study gets started, initiated, what is the standard of care for these patients, and has it been changing at all since the start of the trial?

Right. So I would say that they are hit again, this is a mixed bag; it depends on what center you are at. The busy centers in the U.S. and across some of the countries in Europe looks to put patients on this trial first, because I think that they thought and now the data shows that this is very beneficial to the patients. So, I would say that, that would be placement on a clinical trial is currently probably the first choice of most busy centers. If you are seeing the occasional metastatic ocular melanoma deliver, I can imagine that they are putting the patients on single agent immunotherapy or combination immunotherapy, which I have commented previous on the call that I don’t feel, and the literature will support that it is not very efficacious versus TACE or Y90. There really is no documented and go-to-standard of care. If you see a medical oncologist, they might put you on immunotherapy. If you see interventional radiologist, they might start pushing you towards Y90 or chemoembolization. Hopefully, if this is approved, everybody will start telling the patients that PHP is the best first treatment choice.

And one last question for me, other than the hurdle of learning a new procedure, is there any other reason why patients could be ineligible to go through that PHP?

So, I mean I can give my answer and maybe Johnny can comment as well. We will do a PHP on patients with limited extrahepatic disease, and as you probably know, ocular melanoma predominantly and sometimes solely metastasizes to the liver. It’s different than cutaneous melanoma, which metastasizes everywhere. Ocular melanoma has some sort of predilection to metastasize to the liver. So, concentrating your treatment at the liver at the site of predominant or sole metastatic disease makes sense. If the patient has extrahepatic disease that we can’t control with stereotactic radiation or surgical excision subcutaneous nodules here or there, then they are probably not the best candidate for PHP until further evidence supports combining these types of procedures, PHP with other systemic therapies. Obviously, we are not even near there yet. But patients who have a certain disease burden outside the liver, it wouldn’t be the best therapy to start concentrating their treatment just in the liver. Patients with underlying liver dysfunction are also not the best candidates for this type of procedure. So, if you have underlying cirrhotic liver, Child-Pugh class B or C, these are patients that aren’t the best candidates for the procedure. Obviously, any known allergies to contrast dye because that’s what we use to make sure that the catheters are in the right place to make sure there is no leak in chemotherapy, allergies to latex products. All these are contraindications to performing the procedure, but these aren’t extremely common issues. Obviously, the extra hepatic disease is the one sticking point with these patients that usually makes them not eligible for a percutaneous hepatic perfusion.

I would just stress, he has mentioned most of the points and the liver tumor burden, the Child-Pugh score and so on. We did incorporate the majority of those that he spoke as inclusion-exclusion criteria in the trial. I think those selection of patients has borne out in the safety profile that we have seen and the increase in the number of cycles that we have seen in this trial versus the previous trial. So, I think there was a previous question asking about the average cycle, and currently, it’s between 4 and 5, but the majority of patients, among those that were treated ended up getting 6 cycles. In the previous trial, it was below 3, was the mean or average. Part of it is the safety requirements that we put in place for these patients, and tumor burden being the large one. We have capped it at this trial at 50%. We know that in the European experience they have taken that up a little bit higher. But we do require that the patient has some functioning liver parenchyma to undergo the treatment and recover for subsequent treatment. One of the requirements in the trial was that the patient returned to baseline levels in terms of the hematological and liver enzymes prior to being able to receive subsequent treatments. The fact that these patients were able to receive 4, 5, and 6 treatments shows that the patients could recover and the adverse events were transient in nature. But I think I hope that that answers your question on the criteria for our treatment.

And Johnny, I will add if they go on to 4, 5, and 6 treatments, then obviously, the procedure is working, right, because we are not going to continue perfusing their liver with evidence of progression in the liver and/or progression outside the liver. So, obviously, it’s a testament that the procedure works if you are getting in 4, 5, 6 treatments at every 6 to 8 weeks. If you extrapolate that, that’s up to 48 weeks from the start of identification of the patient as a candidate.

Alright. I want to chime in because I am getting quite a few texts and tweets, etc., on the data. There is, I guess, a theme out there that because the confidence interval on the ITT population, the ORR fell to 12%, that we have failed the trial. I really caution any investors out there who are running with that, that’s incorrect. If necessary, I will put something up on the website or another release to make sure that this is clear. As I said before, when the power calculation was done for the trial, when it was amended to a single arm, we looked at what the upper bound was based on a meta-analysis, and the upper bound was 0.083. So, if the lower bound of the HEPZATO overall response arm is greater than the upper limit of that 0.083, then we have met the endpoint. The calculation determined that a 21% point estimate would beat that, and that’s indeed what occurred. So, for those of you focused on the left-hand side of the lower bound, the lower bound, we had this – we had to pass 0.083, and 21% for the power calculation that was done when the trial was amended did that. I will continue to see that this is a misconception out there, and I will put something out publicly beyond just this call, which might catch some people flat-footed. I just want to make that clear to investors regardless of what side of the fence you are on; I just want to make sure everyone is playing with the same set of facts. With that, I think we are done with questions. So I will just pick it up very quickly from here. I wanted to briefly but use some other recent events and touch upon strategic priorities, which I shared during our last quarterly call. Those priorities are to complete the FOCUS trial and submit an approvable NDA, prepare for the commercialization of HEPZATO kits in the United States, raise the awareness of Delcath in the investor community, and then lastly, prioritize additional indications for HEPZATO kit and then design and execute a development plan for the higher value indications. The preliminary results released today should greatly bolster the confidence that we will be able to compile an NDA package, which supports a favorable benefit-risk profile, and we look forward to completing the trial and discussing the results with the FDA. We are still projecting submission of the NDA to the FDA in the first quarter of 2022, which would still allow for a 2022 launch. Given the COVID restrictions and the backlog monitoring that occurred due to the company’s earlier financial difficulties, the primary gating item to our submission is still monitoring. I must stress the team’s focus is not primarily speed but quality. Our goal is not to submit an NDA as quickly as possible to submit, but to submit an approvable NDA in a timely manner. Our second priority, preparing for the commercialization of HEPZATO in the United States, is the focus of significant activity. In December, Kevin Muir joined the company as Vice President of Commercial Operations. In his most recent role, Mr. Muir was Director of Sales for the Embolics Interventional Oncology business unit of BTG, where he played a key role in growing that business from $40 million to $180 million, at which time BTG was acquired by Boston Scientific. Kevin’s proven ability to both build commercial teams and introduce novel technologies into the marketplace will be a critical asset as we prepare for the launch of HEPZATO upon anticipated FDA approval. This hire is another important step in Delcath’s transition from a development to a commercial stage company. In addition, we continue to work with a well-known consulting firm to lay the groundwork to ensure appropriate patient access and reimbursement and are actively recruiting to build our market access team. Our third priority is to raise awareness with the investor community and ensure the company stays on solid financial footing. In December, we completed an underwritten public offering of 1.7 million shares of common stock at $13.25 per share yielding gross proceeds of about $22 million. The straight common deal was well oversubscribed and included numerous fundamental healthcare funds. As we work to expand the set of investors backing the company, we are erring on the side of transparency in detail, as I hope was evidenced by Dr. John’s presentation. Our fourth strategic priority is the prioritization of additional indications for HEPZATO and the subsequent design and execution of development plans for the higher value indications. Earlier this year, we initiated a consulting engagement to select a portfolio of follow indications and on track to complete this work by mid-summer. In addition to new indications, we are eager to learn whether HEPZATO could improve the efficacy of checkpoint inhibitors. We look forward to the results from an ongoing single-center IIT Phase 1/2 study, the study conducted by Dr. Mark Burgmans and his colleagues out of the Leiden University Medical Center, the Netherlands, investigating combination therapy of Delcath hemostat hepatic delivery system with checkpoint inhibitors in order to better control both hepatic and extrahepatic disease. Results of trials investigating the efficacy of checkpoint inhibitors alone have been disappointing in patients with ocular melanoma metastases. Tumor lysis and necrosis induced by CHEMOSAT could potentially provoke damage and release that may stimulate a cancer-specific immune response and increase the efficacy of checkpoint inhibitors. In summary, I believe the preliminary results released today support that HEPZATO can provide meaningful clinical benefits to ocular melanoma patients, if and when approved, and provide evidence that HEPZATO was a high-value platform with the potential to address multiple cancer indications of high unmet need. Finally, we are on track to submit an NDA in the first quarter of 2022 and are preparing for commercialization. In closing, I want to thank you again all of you for taking your time to listen and ask questions this morning and a special thanks to Dr. Zager for taking time out of his busy schedule to join us. This is an exciting time for Delcath, and I look forward to updating all of you as we work towards making HEPZATO available to all patients who could benefit from its use. Thank you very much for your time today.

This concludes today’s conference. You may disconnect your lines at this time. Thank you for your participation.