Hello everyone, this is Dr Bishal Gyawali from Queen’s Universtiy, Kingston, Canada and it’s my big pleasure today to bring you the highlights of the ESMO 2022 annual meeting. I’m not there in person this time, I’m participating virtually, but this is just a quick note to say that physicians and researchers from low and middle income countries, there is so much inequity in access to cancer care and outcomes that we always talk about, but this is just a small snapshot that the reason I could not be there in person was because I did not get my visa in time. So just an example, and I think I am one of the privileged researchers among those who come from low and middle income countries, but just to give an example of how much barriers there are. We talk about equity, we talk about equality but even participating in a conference becomes a big deal for LMIC researchers. It’s not only about the cost of participation and the cost of visas and the cost of all these travel arrangements but also the mere fact that even when you are speaking at a conference, to get a visa in time to speak at the conference is not straightforward. Sometimes we forget the privilege of not having to go through those problems but I just mean to give a shout out to all the researchers from LMICs, so busy in their work, who have a huge load of patients to take care of who do not have any resources or any support or funding to conduct all the research that they have been doing and then who go through all the hassle to travel and present at conferences. So I just want to give a big shout out to all of them.

Now, starting with our usual ESMO roundup; I was very happy and surprised, pleasantly surprised, at some of the studies that were presented at this year’s ESMO meeting. I think we can broadly categorise the major highlights from this ESMO meeting into two major buckets. One is trials in the early setting, early cancer setting, neoadjuvant and adjuvant trials, and the other is trials that are looking more or less towards the approach of de-escalation of treatment. Both of these are very important. I have written a number of articles about why it is important to de-escalate treatments, why we are always over-treating our patients and why we might be able to safely de-escalate some of that and achieve similar outcomes. I have written about the imprecision of immunotherapy – how we have no good biomarkers to select for immunotherapy; why potentially we are treating too many patients for too long a time at too frequent intervals with immunotherapy without a solid rationale. I was happy to see that some of those cautions were attempted to be answered at this year’s ESMO meeting.

So, first, let me start with the early setting. In the early setting first a couple of negative trials, negative in the sense that these trials did not meet their primary endpoint, but quite important and informative trials. The first trial I want to discuss today in the early setting is the CANOPY trial in lung cancer. Again, as I mentioned, this was a negative trail but this was an important hypothesis to be tested because inflammation is supposed to be an important cause for development and progression of lung cancer; in the parthenogenesis of lung cancer inflammation plays a key role. This trial showed that the use of canakinumab, an inflammatory marker, did not improve outcomes so that hypothesis has been tested and it failed, which is fine. That’s why we do trials. 

But, connecting with lung cancer, it was also very interesting, even for a non-basic science researcher, a clinical researcher like me, to watch the Presidential Symposium from Dr Charlie Swanton about the pathogenesis, the mechanism, why non-smokers also get lung cancer and the role of air pollution into that. That was quite important and ground-breaking because I’m pretty sure almost all oncologists get that cancer when we diagnose someone with lung cancer who is a never smoker, who has never smoked, the first question they ask us, ‘I have never smoked, why did I get lung cancer?’ I think at least we have some mechanism to understand and address that question now.

Moving on to the list of negative trials in the early cancer setting, there were a number of negative trials in renal cell cancer. The first is the adjuvant atezolizumab trial, the IMmotion010 – a very clearly negative trial. Then there was the perioperative, the PROSPER nivolumab trial, again clearly negative. Then there was the CheckMate 914 adjuvant nivolumab ipilimumab. All these trials were negative and this was in contrast with the KEYNOTE-564 trial of pembrolizumab in the adjuvant setting in renal cell cancer which was positive for disease free survival, overall survival data is still immature. 

However, there are several questions to be asked when you get one positive trial in a similar setting and three negative trials in the same setting. I think we need to ask and think carefully about why this is so. Is it true that there are differences between PD-1 and PD-L1 inhibitors? Is pembrolizumab truly different from nivolumab or is it something else that’s going on? There was also the talk of in the neoadjuvant nivolumab trial there was also a discussion about including surgery as a disease free survival event and whether that was appropriate. I think that is appropriate; in neoadjuvant trials you have to include surgery as a DFS event because unable to undergo surgery is an important outcome and if we are trying to test a neoadjuvant approach we need to be able to look into all the downstream consequences of our plan to do that intervention. So it is important to include surgery as an important DFS event. 

More importantly, why all of these trials – neoadjuvant, adjuvant trials in renal cell cancer – are negative and KEYNOTE-564 alone is positive. I do not have clear answers to that question but we need to also look into what percentage of these patients did get immune checkpoint inhibitors when they relapsed. If at the time of metastasis or if at the time of relapse they did get immune checkpoint inhibitors and achieved similar survival, then that could easily mean that we do not probably need to give adjuvant immunotherapy, we could resolve it to be used at a future time point in some patients, rather than using it up front in all of the patients as adjuvant.

Continuing with the negative immunotherapy trials, KEYNOTE-412 in head and neck cancer, event free survival of 0.83, not significant, overall survival also not significant. They tried to look into different subgroups – CPS score >1, CPS score >20 – not significant in any of the subgroups. Locally advanced head and neck cancer. Again, among the uses of immunotherapy in several cancer types, in head and neck cancer it has always been not as straightforward as in lung cancer, renal cell cancer, melanoma. In head and neck cancer, even in the advanced setting, the positive pembrolizumab trials, they were actually negative in the first couple of interim analyses and finally it turned out positive by a very small number of events changing the outcomes. Here in the locally advanced setting this trial is negative. So I don’t think immunotherapy… we need to look at real world outcomes to see if immunotherapy has made much of a difference in head and neck cancer but I personally am not very… I am disappointed with the performance of immunotherapy in head and neck cancer – it has not been as big a breakthrough for our patients with head and neck cancer as it has been for other tumour types. 

So these were the negative trials in the early disease setting, now let’s talk about some of the exciting positive trials in the early disease setting.

Neoadjuvant immune checkpoint inhibition in locally advanced MMR-deficient colon cancer The NICHE-2 study

Let’s talk about the NICHE-2 trial in colorectal cancer with mismatch repair deficient protein, or MSI high. We know that MSI status is an important biomarker for efficacia for immune checkpoint inhibitors. In the NICHE-2 trial, this was an academia-led trial, they tested neoadjuvant checkpoint inhibitor in locally advanced colorectal cancer. They included important high risk patients because we know in stage 3 colorectal cancer these patients have a high risk of relapse. We give them adjuvant chemotherapy and, despite getting adjuvant chemotherapy, a substantial percentage of the patients do end up relapsing. So there was an important need for newer treatments that could improve the risk-benefit equation for this group of patients. 

In the NICHE-2 trial they enrolled patients with stage 3 cancer; 74% of them had either T4 or N2 and that meant these are really high risk patients. More than half of them had both T4 and N2. The beauty of this trial was in its simplicity – they gave just two cycles of neoadjuvant checkpoint inhibitors, the first cycle was nivolumab plus ipilimumab and the second cycle was just nivolumab alone. So this is just four weeks of treatment, just two cycles. Then they looked into the outcomes. The first one was what percentage of the patients can successfully undergo surgery, and that is 98% so almost all patients could undergo surgery – that four weeks of delay did not cost them the surgery, which is an important outcome. Then they looked at outcomes after surgery and they found that 95% of the patients achieved a major pathological response, that means that the residual cancer was less than 10% of the original volume. This was quite remarkable, 95% major pathological response, and complete pathological response was achieved in 67% of the patients. There have been no disease free survival events so far with a median follow-up of 13 months. 

We saw a remarkable result with dostarlimab in rectal cancer, that was presented at ASCO, in which all the patients had achieved a complete response and they could safely avoid surgery. Here in colon cancer a similar population, an MSI high population, we see 95% major pathological response, 67% pathological complete response and no DFS events so far. This is quite exciting; this are very important results. Of course, with any good trials there are more questions than answers, such as do these patients need adjuvant chemotherapy after all? Could it be only nivolumab instead of nivolumab plus ipilimumab for the first cycle because ipilimumab adds to some toxicities? Why two cycles? But the other question could be is surgery even necessary, but I think that would be a little too premature to forego surgery altogether here. Even in rectal cancer we do not have much follow-up data; we had just a handful of patients and unless we have mature follow-up data, foregoing surgery might be a little more risky without having solid evidence. But in colon cancer as well we have to answer those questions in future trials but these are quite exciting results. If access is not an issue then neoadjuvant checkpoint inhibitor in high risk, locally advanced colorectal cancer should be practice changing. I was very excited with those results.

Neoadjuvant versus adjuvant pembrolizumab for resected stage III-IV melanoma (SWOG S1801)

Moving forward with another early setting trial with positive results, that is melanoma, a neoadjuvant versus adjuvant pembrolizumab trial. This was very intriguing for me because there are some rationales, there are some explanations why immunotherapy might work even better as neoadjuvant rather than adjuvant but so far I had not been convinced with that. But these trial results are striking and leads us to think is that actually true and is that true in other disease settings as well; is it only melanoma? It opens up a can of questions. The results are very striking here, using pembrolizumab, just three cycles, before surgery and 15 cycles after surgery versus the usual practice of adjuvant pembrolizumab in melanoma. They achieved a hazard ratio of 0.58 and a two-year event free survival of 72% versus 49%. We need to see whether this turns into overall survival as well. But if you are going to use pembrolizumab in the adjuvant setting anyway, you might as well use three cycles before the surgery and achieve better event free survival. 

But the bigger question is why is this happening, a, and ,b, as we saw in NICHE-2, just two cycles of neoadjuvant checkpoint inhibitors were enough to give such impressive results one might very well ask are those 15 cycles after surgery even necessary? Will just those three cycles before surgery be enough? We need to look into the granular details of this data – what percentage of the patients got after surgery treatment. If there were a good number of patients who got only the neoadjuvant phase and did not get the adjuvant phase treatment and still achieved those excellent outcomes then it could meant that we could give just 3-4 cycles before surgery and stop the treatment. As I have been saying, probably we are giving too much treatment, too much immunotherapy treatment, to our patients so there might be several opportunities for de-escalation, especially with these exciting neoadjuvant results. Maybe one year, two years of adjuvant therapy that we are giving with immunotherapy nowadays are probably not necessary, probably we could achieve similar or even better outcomes by giving just three or four cycles before surgery. So these are exciting questions but, again, we will have answers to those only after doing clinical trials. That’s why trials are so important.

That’s the theme of early setting trials. The other major theme, as I was saying, was the theme of de-escalation of treatment.

Nivolumab (Nivo) plus ipilimumab (Ipi) 6-months treatment versus continuation n patients with advanced non-small cell lung cancer (aNSCLC): Results of the randomised IFCT-1701 phase III trial

Again, let’s start with lung cancer. The IFCT trial, an academic-sponsored trial, again asking quite relevant, important questions of should we give immunotherapy for ever until disease progression or toxicity, or can we stop after a certain number of cycles, the so-called stop-and-go strategy. In this case, unfortunately, I don’t know why, the trial started with nivolumab ipilimumab as the immunotherapy treatment for lung cancer when in fact pembrolizumab is the most commonly used in this setting. That’s the reason why they had to stop the trial early, because it did not get registration in the EMA so they could not continue the trial which is unfortunate. But in this trial when they stopped the treatment after six months versus continuing it until disease progression, they actually saw remarkable results. It was quite encouraging for the stop-and-go strategy but, as I mentioned, the trial had to be stopped early so we cannot infer any definitive conclusions. But what is encouraging is the feasibility of doing such trials and the possibility that in the future we will do such trials of stopping immunotherapy treatment after certain cycles rather than continuing it for ever. Maybe we can get some positive results and maybe we can spare our patients with the toxicities of treatment, the time of toxicity, the financial toxicity, all sorts of toxicities associated with continuation of treatment beyond what is needed.

Comparison of abiraterone acetate and prednisolone or combination enzalutamide + AAP for metastatic hormone sensitive prostate cancer starting androgen deprivation therapy: Overall survival results of 2 randomised phase III trials from the STAMPEDE protocol in prostate ADT therapy

Then the STAMPEDE trial in prostate cancer – no big cancer congress feels complete without one major presentation from the STAMPEDE team. I’m super-impressed with the STAMPEDE trial team and I quote this example everywhere. The STAMPEDE trial is the best example of any contemporary cancer trial going on in the world; they have been delivering results one after another. Their contribution to prostate cancer is matched by no other single trial in any other tumour type. So, again, in this ESMO meeting the STAMPEDE trial presented the data for a combination of abiraterone plus enzalutamide in relation to androgen deprivation therapy. We know separately that ADT plus abiraterone is effective, ADT plus enzalutamide is effective, also the triplet results were shown last year – ADT plus docetaxel plus abiraterone. But in this ESMO they presented the data for ADT plus abiraterone plus enzalutamide, is it better than ADT plus abiraterone alone. The answer is no so, again, it can fit into our overall theme of de-escalation. Not that we have been doing abiraterone plus enzalutamide but they have shown that it’s not an effective strategy, we just add toxicity. So we can stay with ADT plus abiraterone and there is no added advantage of combining enzalutamide on top of that.

Adjuvant hyperthermic intraperitoneal chemotherapy in locally advanced colon cancer (HIPECT4): a randomised, phase 3 study

Then moving on to colorectal cancer, there were a couple of interesting trials. One is the trial that tested HIPEC, a regime of HIPEC with intraperitoneal mitomycin C in colorectal cancer. I want to add a quick note about the HIPEC strategy. It has been tested in several tumour types that are prone for peritoneal metastasis but I don’t think it has had any solid evidence in any of the tumour types. But I know that several centres around the world are pursuing a HIPEC strategy. I think we need more trials and unless I see any positive trials I think HIPEC should always be considered an investigational treatment, it should not be considered the standard of care. The reason I say this is I have seen patients in several parts of the world, in low and middle income countries and also in high income countries, think of HIPEC as a pretty big deal, as if it is going to cure their cancer and spending all their time, money and resources in pursuing HIPEC therapy. But it’s a note of caution – we have seen a number of negative trials for HIPEC in several tumour types. In this ESMO meeting for colorectal cancer we saw that it improved local regional control of disease but it did not improve disease free survival overall or the overall survival. So this is a negative trial as well.

FOLFOX plus nivolumab and ipilimumab versus FOLFOX induction followed by nivolumab and ipilimumab in patients with previously untreated advanced or metastatic adenocarcinoma of the stomach or gastroesophageal junction: Results from the randomised phase II Moonlight trial of the AIO

Continuing with colorectal cancer, one interesting trial does not fit into the theme of de-escalation per se but one interesting trial was testing nivolumab in gastroesophageal cancer in combination with FOLFOX as sequential versus parallel. Right now we give FOLFOX plus nivolumab but the authors tested whether giving FOLFOX first followed by nivolumab would achieve similar outcomes. That is an important strategy because if you can give FOLFOX only and then nivolumab probably you don’t add to the toxicity and then it’s much easier to manage. So that was a good question to ask but unfortunately the answer was not what we had hoped. So the answer is that we have to give FOLFOX plus nivolumab and FOLFOX followed by nivolumab does not necessarily achieve similarly good outcomes. 

FOLFOX/FOLFIRI plus either bevacizumab or panitumumab in patients with initially unresectable colorectal liver metastases (CRLM) and left-sided and RAS/BRAFV600E wild-type tumour: Phase III CAIRO5 study of the Dutch Colorectal Cancer Group

The other colorectal cancer trial was the CAIRO5 trial in which for the left-sided RAS wild colorectal cancer they again tested bevacizumab versus panitumumab in combination with chemotherapy for patients with liver metastasis. They found that the resection rate, R0, was similar, PFS was similar; the response rate was higher with panitumumab but other outcomes were mostly similar. This is another trial to add to the body of several other trials that have been testing bevacizumab versus EGFR antibodies in this setting.

Pancreatic cancer trial – Evaluation of gemcitabine and paclitaxel versus gemcitabine alone after FOLFIRINOX failure or intolerance in metastatic pancreatic ductal adenocarcinoma: Results of the randomised phase III PODIGE 65 – UCGI 36- GEMPAX UNICANCER study

In GI cancer there was one interesting trial in pancreatic cancer that is gemcitabine plus paclitaxel second-line treatment. We know gemcitabine plus nab-paclitaxel as first-line treatment and there is also FOLFIRINOX in first-line treatment and usually the practice is for patients who are fit we use up-front FOLFIRINOX and then the patient progresses. Depending on where you are practising and the approval and the reimbursement system, you usually run out of options because you might want to use gemcitabine plus abraxane but it gets funded only in the first-line setting, such as in Canada, so you can’t use it in second line. 

We have always been thinking is nab-paclitaxel different from paclitaxel in terms of efficacy outcomes because we did not test gemcitabine plus paclitaxel, we tested directed gemcitabine plus nab-paclitaxel. But now we’ve got a trial of gemcitabine plus paclitaxel. This is the second-line setting and unfortunately even in second line it did not improve any survival as compared to gemcitabine but there was some difference in PFS. More than anything it allows people to use gemcitabine plus paclitaxel in pancreatic cancer after progression on FOLFIRINOX but, again, we have to be very careful because survival has not been proved and it’s just improvement in PFS in the second-line setting of pancreatic cancer. So if a patient has progressed on first-line treatment and in the second-line therapy we need to be very careful to evaluate whether or not they need this combination treatment. Especially when we have seen that survival is negative we might just be adding toxicity than adding some more days of toxicity-filled treatment without improving any survival. So that’s a word of caution there.

Primary results from the phase III LEAP-002 study: Lenvatinib plus pembrolizumab versus lenvatinib as first-line (1L) therapy for advanced hepatocellular carcinoma (aHCC)

Continuing with our GI theme, let’s talk about a couple of trials from hepatocellular cancer as well. Because this ESMO was a unique meeting because you don’t see three or four hepatocellular cancer trials presented at one cancer congress. In this ESMO congress we had several hepatocellular cancer trials presented, let’s just start with a negative trial; it fits into the theme of de-escalation. Lenvatinib plus pembrolizumab versus lenvatinib plus placebo did not improve PFS, did not improve OS. The median overall survival in the placebo arm was 19 months which is remarkable for hepatocellular cancer but it also tells you how highly selected these patients are that go into hepatocellular cancer trials. I have written about it in the past: in the real world patient outcomes are far worse but if you look at the clinical trials, if you look at the placebo arm, most of the patients in the placebo arm are doing better than what our patients in the treatment arm do in the real world. 

But this also asks the question of we have had the positive atezolizumab plus bevacizumab trial in hepatocellular cancer, and we can make this argument with the renal cell cancer adjuvant trial as well, is that once we have one established immunotherapy, until what time can we continue to test these trials with an outdated control arm? Because we have clearly shown that adjuvant pembrolizumab is better than placebo in renal cell cancer, we have shown that atezolizumab is a standard treatment, it improves survival substantially in first-line hepatocellular cancer, but we still are continuing to use placebo as the control arm of renal cell cancer adjuvant trials, we are still using sorafenib and lenvatinib as the control arm of first line hepatocellular cancer trials. 

Camrelizumab (C) plus rivoceranib (R) vs. sorafenib (S) as first-line therapy for unresectable hepatocellular carcinoma (uHCC): A randomised phase III trial

As I mentioned, the other trial of camrelizumab plus apatinib, again in first line hepatocellular cancer, was tested against sorafenib, an outdated control arm. They showed that this newer combination significantly improved survival but also toxicities, added toxicities. One thought on this trial is that there is some hope that this regimen might turn out to be cheaper than atezolizumab plus bevacizumab because these are, as far as I know, molecules developed in China and they are trying to compete in terms of cost of the drug. So if that brings the overall cost of treatment down for first line hepatocellular cancer then that could be a win for the patients and for the community.

Final analysis of RATIONALE-301: Randomised, phase III study of tislelizumab versus sorafenib as first-line treatment for unresectable hepatocellular carcinoma

Then there was the trial of tislelizumab versus sorafenib, again sorafenib as the control arm, I’m not happy about it, in first-line hepatocellular cancer. But this is asking one interesting question that the atezolizumab trial tested atezolizumab plus bevacizumab and we do not know what is the added contribution of bevacizumab there – would atezolizumab alone have been effective? So this trial actually tested monotherapy checkpoint inhibitor versus sorafenib and unfortunately the overall survival was not improved, it was only non-inferior. There is no point in using a checkpoint inhibitor that is just non-inferior. So those are the major trials.

DeFi: A phase III, randomised controlled trial of nirogacestat versus placebo for progressing desmoid tumours

A couple of extra notes, one about the sarcoma trial. In desmoid tumours the DeFi trial of a gamma secretase inhibitor, nirogacestat, a very difficult to pronounce name. A special note about this trial because desmoid tumour is not a very common tumour, it’s a rare cancer. First, kudos to the investigators, it shows that you can do a randomised trial even in such rare tumour types. Second, they measured not only response rate and progression free survival but also measured quality of life and kudos for that. This is an important point because desmoid tumours are not always life-threatening survival events. This could be one of those settings where overall survival is not a good endpoint because this does not always lead to survival events; this cancer is not lethal but it can be pretty disfiguring, it can cause quality of life complications, it cause pain. So measuring quality of life is an important endpoint here. Response is also an important endpoint here and they showed that the response rate difference is 41% versus 8% with placebo.

I want to highlight this point because there is an 8% response rate even with placebo. That’s the reason why we need to measure even response rate. Even if we are giving accelerated approval based on response rate it’s always important to measure it in a randomised trial fashion because we have had certain drugs that have been approved on the basis of 13-14% response rate. 14% response rate, without a randomised trial you would not know whether that is a really impressive response rate worthy of approval or not. As I mentioned, in this trial we saw 8% response rate with placebo alone. Quality of life was improved so this was an important trial.

Treatment with tumour-infiltrating lymphocytes versus ipilimumab for advanced melanoma: Results from a multicentre, randomised phase III trial

The other remarkable trial in terms of scientific breakthrough was the trial of tumour-infiltrating lymphocytes versus ipilimumab in the second-line setting for metastatic melanoma. Substantially improved progression free survival with a hazard ratio of 0.4 but OS is so far negative with a hazard ratio of 0.83; it’s not mature yet. Quality of life improved but the administration of tumour-infiltrating lymphocytes is not easy, there are a lot of logistical challenges to it, so it is remarkable from a scientific breakthrough point of view and hopefully it will open up other avenues. But until now, given the negative overall survival and the logistical challenges, I don’t think this can become immediately practice changing. We need to wait for the survival results and figure out how to overcome those logistical challenges so that institutions can deliver it easily and quickly.

Thank you very much for listening; that concludes my summary for the major trials from ESMO 2022. I know I haven’t talked about the gynaecological and breast cancers; I will make a separate video for that. But overall I was happy with these trials that fit nicely into the de-escalation strategies as well as several trials in the early setting. Because the early setting is where we can achieve cure, where we can try to pursue that elusive goal of making our patients with cancer cancer-free. We have had several disappointing results but those are a part of the learning process. But we also have had some quite encouraging results like the NICHE-2 trial and the neoadjuvant trial in melanoma. Our field is changing so that the fraction of patients who get cured keeps increasing.