DS-8201 has a PDUFA Date

Yes! The date is F1Q:20, so I might be getting a prolonged OS for Christmas!


Daiichi announced today that they’ve been granted FDA Priority Review for their highly worthy DS-8201, which looked promising out of the gate and recently demonstrated a 59.5% response rate in HER2+ metastatic breast cancer patients who had previously been treated with T-DM1. Those results, from a 118-subject Phase 1, were published in June. The study also reported some total insanity, like a 20.7-month duration of response (DOR) and a 22.1-month PFS. It’s not comparable given the phases, designs and number of subjects enrolled in this phase 1 versus EMILIA, but T-DM1 had a phase III PFS of around 9-10 months, so DS-8201 is likely to displace that agent as the second-line treatment of choice in this setting (first-line being taxane + Herceptin + Perjeta).

The data above are old news, but the press release says the submission package also includes pivotal phase 2 DESTINY-Breast01 data, which “validated” the phase 1 results and which no one has seen yet, though they note that it will be presented at San Antonio. DESTINY 1 looks at a similar population (T-DM1 pre-treated MBC), but its 230 subjects were randomized to varying dose levels. Primary endpoint is ORR, so it would be shocking if this result isn’t pretty compelling, considering what we’ve seen so far. 

Like T-DM1, DS-8201 is an ADC where Herceptin is conjugated to a chemo agent; in DS-8201’s case, it’s a TOP-I inhibitor, similar to irinotecan.

Where did we leave off?

Was it the brain mets? We’ll start with the brain mets.

There were a bunch of them. A dozen? A couple were giant. I had whole brain radiation, which I always swore I wouldn’t do out of fear of cognitive decline. Six months out, I can still tie my shoes and spell my name, so I guess it’s turning out fine. 


When the WBR was over, I did what any reasonable person with no access to tucatinib and looking to avoid the Maintenance Nerlynx would do: I hopped on a plane to Germany. (Before you judge me for my rash leap to dangerous, unproven treatments administered in rogue foreign clinics, rake Pubmed for some rigorous efficacy data on FDA-approved brain mets interventions. I’ll wait.) 

One of the reasons I was excited to come back to the blog was so I could tell this story. My German doctor almost made me cry the day I met him, in the most unexpected way.

He was walking me through treatment options in his office, some EMA-licensed, some not, many occupying a regulatory gray area for which we don’t really have a U.S. analogue. The rules around the manufacture and administration of these treatments seemed a little vague, and my worry was that the treatments themselves would end up supply-constrained (what would be more frustrating than having one dose of a treatment and then never being able to secure a follow-up dose?), or that the clinic itself would be shuttered for some nonsense violation, and I’d lose access that way. Note that I wasn’t worried about safety, which is a fun consequence of the brain mets. What can hurt me now but being afraid?

My doctor assured me that it wouldn’t be a problem. He elaborated, and some of this (particularly the legal circumstances and terminology) may have been lost in translation, but the gist of it was this:

There was a court case in Germany where a physician was charged with murder for falsifying lab results and other medical records for a patient in need of an organ transplant, making the patient appear sicker than he was. For this reason, the patient was moved up the transplant list and received the transplant while other patients awaiting organs died. The falsified documentation was uncovered, and the physician was charged and found guilty. On appeal, a higher court reversed the judgment, deciding that the doctor was responsible for his patient, not every sick person in Germany.  This set a precedent that gave physicians a lot of latitude in patient care. He’d go on to add that it applied to unproven or unlicensed therapies if the doctor determined that the benefits outweighed the risks, which he made evident to me was a pretty low bar given the brain mets.    

All I heard, loud and clear, was My only obligation is to you.

I almost burst into tears. What is this blog other than the hope that a doctor would say those words to me? I love trashing scummy companies as much as anybody, but why would any of this bother me so much if I weren’t clinging to that ideal? In that office, suddenly, my guard was down and I relaxed; I would try anything, pay anything, keep any secret, now and later and long after I’m gone (let my husband deal with that one; tell them the nachtkrapp got me), because for once the transparency existed where it mattered and not where it didn’t.

U.S. healthcare is programmed so this scene would never happen, and I would argue that the cost of that is trust. The physician-patient relationship is not a partnership, and it won’t be as long as we continue to cloak the reality that your doctor doesn’t always want what you want. This was not my first experience with ex-U.S. care, and I’m conflicted about those experiences; in the EU I received in many respects higher-quality, more pragmatic care than I could dream of in the U.S., but there was also some great frustration that an American adult who grew up in the candy store of U.S. healthcare can probably never reconcile. But that conversation, in that German clinic, was the clearest validation I’ve seen that the ideal is possible. It can be done. This is how you empower a patient.

Puma Releases NALA Results

 

Puma’s announcement on Monday that they hit one of their two primary endpoints in NALA is one of the more opaque press releases I’ve seen in a while, reporting a statistically significant PFS benefit for neratinib + capecitabine versus lapatinib + capecitabine, an OS benefit that did not reach significance and a statistically significant benefit in a secondary endpoint of “time to intervention for symptomatic central nervous system disease.” The latter endpoint is the one I most want to be meaningful because CNS was always the most compelling use case for this drug, and wouldn’t it be great if neratinib actually helped someone? Even when you squint, though, there’s not much to be encouraged by in these results, and this is a company press release. 

NALA is a phase III randomized study concocted on the whim of becoming third-line SOC in HER2+ metastatic breast (after THP and T-DM1), despite the fact that DS-8201a and, more relevantly for CNS mets, tucatinib have made considerably stronger moves to claim that spot. (DS-8201a is going after second-line, and we know nothing about the drug’s activity in CNS disease; tucatinib has focused on CNS.) There is nothing in Monday’s press release to suggest that neratinib has a credible claim to the third-line spot; a PFS edge over lapatinib + cape is nothing to write home about, and most patients in the 3L setting would likely prefer a trial or a more compelling Herceptin cocktail (more on the absence of Herceptin below), over neratinib + cape. This is not going to be the standard third line for visceral disease. That means we’re left with a CNS opportunity, which is a sad slice of the HER2+ market opportunity and, importantly, not all that different from what’s available now via guidelines. Since all we’ve seen from NALA so far are delays attributed to a slow rate of mortality events, with no info about the patient mix, there’s not much to build confidence in the validity of this benefit. It would be nice to know how many patients are in the CNS analysis, especially since the time to CNS intervention measure is just a secondary endpoint. There might not be anything here, and until we have tucatinib, Tykerb is just so much easier. Plus, in regular clinical use, you get Tykerb with Herceptin. 

The NALA comparator is infuriating, because it’s not real-world standard of care; no one drops Herceptin, and this is one of the reasons I love the Cascadian study; HER2CLIMB’s experimental arm gets tucatinib + Herceptin + cape. The triplet makes sense, has positive signal from a published phase Ib study and aligns with practice in that everyone gets Herceptin. (It’s like they had a market strategy or something, like they were planning to be the TKI of choice in HER2+ brain mets.) We have no evidence that neratinib will perform better than the true SOC, and with the toxicity, why guess? Why put that on the market in such a competitive environment? The evidence we do have trades QOL not for OS, but for imaging results, and it’s just not good enough. 

Actual data was promised sometime in 2019. 

Puma Hobbling Along a Year After Neratinib’s Launch

The Nerlynx needs your help. On their 3Q earnings call, Puma Biotech presented brutal sequential sales for neratinib, tanking the stock, hopes & dreams, etc. 


In millions.

That’s a 3.5% increase from 2Q18 to 3Q18. 

This is not the story for, say, Perjeta, which was initially launched in the small market of first-line HER2+ metastatic breast. A year out from its introduction, in an equivalent period as referenced above, sales increased 34.5% over the previous quarter. Six years later, after having picked up approvals in the adjuvant and neoadjuvant settings, Roche reported Perjeta sales of CHF 2.0 billion ($2.0 billion) for the first nine months of 2018, which is a 24% increase year-over-year. Kadcyla, another HER2+ drug which is still only approved for use in metastatic patients, grew 30.8% between its fourth and fifth quarters on the market and has booked CHF 728.0 ($743.3 million) in sales so far this year, an 8% increase over the prior-year period. Deviating from HER2 for a minute, Ibrance was launched in first-line, HR+ breast and was growing 36.2% a year from launch, only to report nearly $3 billion in sales in the first nine months of 2018, even as U.S. sales are being impacted by increased competition.

This is just an illustration of the growth trajectory that can be expected with a new breast cancer drug managed correctly. Puma’s performance is not normal. They continue to promise that Europe will deliver revenues, but there was one tidbit in the earnings call transcript that I missed before: their proposed label in the EU market is ER+, HER2+ patients, while in the U.S. it’s all HER2+ patients (about 20-25% of the 260k new U.S. breast cancer diagnoses each year). I checked the APHINITY demographics and was surprised to find the ER+/HER2+ cohort was as big as it was – about 64% of patients in that study (compared to 57% in Puma’s own EXTENET) – but they’re still losing a big chunk of the market. Whether the drug will get any traction in the conservative EU market is a different issue, but when Americans are skeptical, what does that tell you?  

Still no news on NALA in the metastatic setting, aside from reiterating that it’s going to be 4Q this year or the first half of next year. No urgency, it seems, though at this rate Kadcyla could be approved in the neoadjuvant setting before we even see NALA, further eating away at the utility of neratinib. For an anecdotal example of what Puma’s extended adjuvant delusion is costing them, my own doctor had a HER2+ metastatic patient with decent systemic control and progressive brain mets. Whatever I think of Puma management, we have one drug in breast cancer that passes the blood-brain barrier. Tucatinib is coming, but at the moment, if you’ve failed Tykerb, neratinib is the only marketed option. My doctor was worried insurance wouldn’t pay for it because of the early-stage label. I said of course they would. It’s in the guidelines.

She had no idea what I was talking about. I pulled up the press release about neratinib being added to the HER2+ CNS guidelines on my phone. I only knew about this because I stalk the company. An actual doctor who needs the drug available had no clue. This is how Puma thinks things should be done. Insurance apparently said yes to the neratinib request on the first go, and the patient is on it now, so you’re welcome, guys. 

The unmanageable diarrhea came up on the earnings call, as it always does, and the keyword remains compliance. Puma assured investors that the specialty pharmacies are sufficiently hounding patients to refill, that they’re offering vouchers for loperamide, and that they’re working to get budesonide and colestipol prophylaxis added to the label (so many drugs for these patients!). What is the rationale? The patient is waffling on refilling their prescription, because, you know, there’s nothing wrong with them, and it’s the free Imodium that persuades her? Like the free hat when you sign up for the low-rent gym in the shopping center parking lot? I can’t think of another management team so out of touch with their own market. How many Puma staff and affiliates are charged with pressuring healthy patients to take unnecessary meds versus reaching the prescribers and patients who could actually benefit from neratinib – and have every motivation to take the drug as prescribed, with the hope of refilling all the way through improved OS? 

Syndax Misses Primary PFS Endpoint for Entinostat

Syndax announced on Thursday that they missed PFS in their P3 E2112 study comparing entinostat + exemestane (Aromasin) to placebo + exemestane in advanced HR+ breast cancer, which resulted in a top-notch corporate quote:

“While the PFS analysis did not show a statistically significant benefit, E2112 was primarily designed to determine whether the combination of entinostat and exemestane could improve OS based on the compelling OS results obtained in the Phase 2b ENCORE 301 trial …”

They didn’t even WANT to hit PFS, okay?

Never mind that there are two primary endpoints in E2112, meaning it’s an and/or situation and the study was not, in fact, primarily designed to assess OS. 

ENCORE 301, the basis on which Syndax received their Breakthrough Therapy designation, is reported here. That study (N=130, randomized 1:1 to entinostat + exemestane or placebo + exemestane) met its primary endpoint, with PFS of 4.3 months in the entinostat arm and 2.3 months in the control arm. OS was an exploratory endpoint, with entinostat associated with median OS of 28.1 months, compared to 19.8 months in the control arm. The enrollment criteria are similar between ENCORE 301 and E2112, except premenopausal patients and prior fulvestrant are both allowed in E2112, and E2112 appears to have a cohort with non-measurable disease. 

It would be unusual for the benefit of an anti-cancer therapy to be better reflected in OS than PFS; this might happen with PD-1/PD-L1s, but that may not be universally agreed, and it’s not likely to be the case for entinostat. Syndax are trying to recast entinostat as an immunotherapy with their checkpoint inhibitor collaborations (oh look, it’s a 10% response rate with pembro), but when I tried the drug a couple years ago, it was just an HDAC inhibitor. And it sucked. 

I was on entinostat for three cycles and got dose reduced after the first and again after the second, which remain my only dose reductions in eight years of nonstop treatment. I was off study for progression at the end of the third cycle, once I’d used up my two protocol-permitted dose reductions without improvement in my labs. Weird how that happened, right?  

Whatever: I was happy to be kicked off, because I have never felt more like dying in my life as on that drug. It was miserable. I felt sick and tired all the time, it wiped out my white cells, red cells and platelets, and my bones felt hollow. It sounds ridiculous, and this was never attributed to the study drug, but they were not like that before. They were totally normal bones before.

The feeling resolved after I went off study, but not before I broke my hand in a freak incident involving a leash that … brushed across my hand very fast. You would think this sort of thing would not cause a spectacular extremity fracture (the surgeon who wired it together later presented it at a conference!), and it could have been a coincidence, but I blame the entinostat.  

The lack of efficacy is the focus here (with a pointed reminder that not all drugs anointed Breakthroughs pan out), and a P3-grade PFS fail rightly overshadows toxicity and my unconfirmed Bird Bones anecdote. I just like the story, lest there be any suspicion that entinostat has “well-tolerated” going for it. In a true stretch of corporate credibility, we’re being forced to wait for E2112 to fail definitively, because Syndax were clear they would keep performing scheduled OS analyses until everyone was dead. Or, you know, they see an OS benefit. Whichever comes first. 

Alpelisib Summits Mt. PI3K

ESMO was a snooze, right? No new drugs out of nowhere, no hotly-anticipated studies knocking it out of the park (cough, IMpassion, cough). Which I guess frees us up to talk about alpelisib.

Novartis announced results from their SOLAR-1 study, which succeeded where buparlisib (also NVS) and taselisib (Roche) failed. In patients with PI3K-mutated, hormone-positive and HER2-negative advanced breast cancer, alpelisib plus fulvestrant achieved a PFS of 11 months, compared to 5.7 months for fulvestrant alone, and an ORR of 36% for alpelisib + fulvestrant, which compares to 16% in fulvestrant alone. These are hormone-positive patients, so we won’t have OS data for another few decades, but these are good numbers. It wasn’t Ibrance/letrozole, which blew out a wild PFS in the first-line setting (notably, the SOLAR-1 patients will have progressed on an AI as well as possibly a CDK 4/6 inhibitor), but it’s good, and the first real validation we’ve seen of PI3K targeting in breast.

This would have been a perfect announcement if not for the AEs. Hyperglycemia was reported for 64% of alpelisib patients (compared to 10% among controls); in 37%, hyperglycemia was grade 3/4. Grade 3 hyperglycemia is defined as blood sugar of 250-500 mg/dL; grade 4 is >500 mg/dL. Which sounds challenging, even in patients who theoretically have bigger problems. These are patients that are poised to live a really long time with a good quality of life. Is any onc out in the community, dealing with real patients, psyched about managing blood sugar of 400? The press release didn’t break out the rate of dose reductions, but what does that inevitability do to efficacy? 

To their credit, Novartis has been fairly relaxed about enrolling diabetics in its PI3K studies. SOLAR-1 excluded type 1, but allowed controlled type 2; BELLE-2 and BELLE-3, the buparlisib studies in HR+ MBC, didn’t appear to restrict based on diabetes or blood glucose, though other, investigator-led studies did. SANDPIPER, the P3 taselisib, excluded Type 1 or Type 2 diabetes requiring anti-hyperglycemic medications. Still, not everyone is going to be represented in the SOLAR-1 population, and there’s a reasonable likelihood that physicians will be prescribing for patients with high baseline blood glucose or less well-controlled diabetes. (And while there’s an argument that metformin and PI3K/AKT/mTOR inhibitors are synergistic, and that by adding metformin to one of these drugs you’d get both protection against a blood sugar crisis and enhanced anti-cancer efficacy, there’s at least some evidence with everolimus that it’s not so straightforward. Though these diabetic neuroendocrine patients who were on metformin when starting evero did better than their non-diabetic counterparts.)

Everolimus is an older, HR+ breast-approved mTOR inhibitor that popped up in a couple of ESMO abstracts (here and here) that aren’t really interesting enough to discuss on their own but that roughly echoed what we know from BOLERO-2, which is that evero doesn’t really care about PI3K positivity. This is notable since this drug for patients with targetable, frequently-mutated PI3K has been one of the prime examples of precision medicine in practice. It’s been offered to me a couple of times based on genomic sequencing (never in a large center, obvs), which I always declined because I was waiting for one of the next-gen PI3Ks. Taselisib turned out to be a disaster, despite the all the strategizing that preceded signing that consent. I think this is a good example of the kind of hold that sequencing has over patients; those mutations feel a lot like capital, and I put a huge amount of effort into deciding when and how to cash it in. Diabetes aside, if alpelisib works, and exclusively in PI3K-mutated patients (there is PI3K-wild-type cohort in SOLAR-1 being tracked for a secondary endpoint), it could help reverse what has felt like a big disappointment in personalized medicine.  

Checking in on Breast ADCs

With Mersana struggling to dig its way out of the pain of a clinical hold, it seemed like a good time to check in on the many antibody-drug conjugates (ADCs) in development in metastatic breast cancer. T-DM1 made it look so easy.

Sponsor Target Payload Status
T-DM1 Genentech HER2 DM1 (maytansinoid) Approved
Sacituzumab govitecan (IMMU-132) Immunomedics TROP2* SN-38 (irinotecan metabolite) Phase 3
Trastuzumab deruxtecan (DS-8201a) Daiichi Sankyo HER2 Exatecan derivative Phase 3
SYD985 Synthon HER2 Duocarmycin Phase 3
XMT-1522 Mersana HER2 AF-HPA (auristatin) Phase 1b
Ladiratuzumab vedotin (SGN-LIV1A) Seattle Genetics LIV-1* MMAE (auristatin) Phase 1
ARX788 Ambrx HER2 MMAF (auristatin) Phase 1
Glembatumumab vedotin Celldex GPNMB MMAE (auristatin) Discontinued at P2b (METRIC)

Development-stage drugs bookended with T-DM1 and Celldex, since they’re referenced in this blog; asterisk if Sponsor claims you don’t need confirmed expression of the marker to benefit from the drug.

Mersana was put under a partial clinical hold in July when a patient died on-study. The hold was lifted two months later with some protocol changes – “increased monitoring”, more limitations on hepatic function for future subjects and a new dosing schedule, with the drug being administered every four weeks instead of every three. The ClinicalTrials.gov status is still active, not recruiting, and they haven’t updated the enrollment criteria or dosing schedule yet. So, as the stock will attest, we’ve heard nothing good about XMT-1522 so far, only bad, but I’m not totally horrified by the safety signal this early. Mostly because this is a heterogenous population in something of a baby basket trial. They’re enrolling the traditional HER2+ MBC cohort, which is, generally, a pretty healthy bunch – but they’ve also got HER2-low MBC, HER2+ gastric and HER2 mutated or overexpressed NSCLC cohorts. The mix as of June was 18 breast patients (HER2 status unspecified), 3 gastric patients (again, no HER2 detail) and 1 HER2-amplified gallbladder patient. There’s some unpredictability there, and likely variation in clinical status.

Would I enroll? No way. The drug is too weird. I don’t want a “novel HER2 antibody”. Mersana says their antibody attaches to a different HER2 epitope than Herceptin and Perjeta; that’s about all we know about it. They’ve also developed a novel binding technique and aren’t using a traditional linker. Too many variables, guys. If you want patients to enroll in a phase 1, at least make them feel like they’re getting some Herceptin. Daiichi and Synthon have an antibody that is virtually identical to trastuzumab, and that strikes me as exactly right; leave the uncertainty to the linker. For what it’s worth, trastuzumab is believed to contribute to the efficacy of T-DM1 – not just due to efficient binding, but by exerting some immune-mediating effects. I’ll take it. Why get creative? We don’t need a more perfect antibody. The benefit is in the payload.

Synthon has been so quiet about their trastuzumab-duocarmycin (alkylating agent) ADC that I was surprised to find that it opened a P3 a year ago in HER2+ MBC, adorably called TULIP. They’re from the Netherlands. They had an ASCO abstract. Preliminary ORR of 33%, and no one died. Their eyes got a little runny. Cool. Sounds good.

Daiichi has been so busy with DS-8201a lately that they probably deserve a separate post (spoiler: am pumped for this drug), and we’ll skip over Ambrx, which no one knows anything about. That brings us to the ADCs that target, well, something. Maybe cancer.

Seattle Genetics is developing what used to be called SGN-LIV1a, which purportedly targets LIV-1. What’s intriguing about this is the evolution of the study design on their P1, which has been open since the Year of Our Lord 2013. It’s progressed from a dose escalation in 50 subjects where positive LIV-1 expression on a new biopsy was required for enrollment to a 300-subject, multi-cohort free-for-all where a new biopsy is required for enrollment. The new biopsy thing kills me; where do sponsors get the audacity to require this? Despite making it a condition of enrollment, they don’t have to pay for it, as it can easily be billed as routine care. More importantly, it’s risky for many patients. Especially considering the risk/reward on P1s: if I can die from unexplored toxicity, you don’t get an invasive fresh biopsy. At any rate, SeaGen has now changed course to say that LIV-1 is expressed almost universally in metastatic breast cancer and reported some early results (25% ORR in a TNBC cohort of 60 subjects). I’m not sure the early responses will stand given Celldex, which also had an MMAE payload; whether the LIV-1 target is viable or not, these patients will have progressed on at least one tubulin binder and likely a couple. Let’s keep expectations low on this one, though they did beat IMMU on opening a first-line TNBC ADC + checkpoint inhibitor study (SGN-LIV1a + pembro) and got into the MORPHEUS study, so they’re definitely part of the conversation.

Immunomedics continues to be irritating. They are relentless in their promotion of their TNBC unicorn that has an SN-38 payload that’s – what is it? – four billion times as potent as irinotecan? Does that mean the patients will survive four billion times as long? IMMU has years of follow-up on these subjects by now, but we’ve only been fed the same tiny nibbles of data over and over; as a reminder, it was a 30% response rate on a cohort of MGH dream patients. This won’t hold up in phase 3, but IMMU got FDA to accept their BLA on Glitter Unicorn-umab on P1/2 data, with a PDUFA date set for January. It’s hard to call whether they’ll get approved or be forced to wait for the P3, but if they do get approved, here’s hoping they go away for a while. With weak data, IMMU should be pretty well trapped in the third-line setting, and we can let more interesting agents (and study designs) sort out earlier lines of TNBC treatment.

With all this activity and so many viable drug candidates, I’m wondering if investors aren’t being more forgiving of Mersana in part because we don’t really need them. We have some near-term approvals coming that will significantly increase the number of ADCs on the market, but the big change since T-DM1 in 2013 is that we’re now fully committed to combination therapy. Targeted drugs with manageable toxicity profiles are making creative cocktails possible, and the results are a lot more compelling than what we’re seeing in the box-checking monotherapy trials that get these agents approved. For instance, T-DM1 performs well on its own, but we’ve already been teased with a 60% response rate when you pair the drug with neratinib. It’s never been easy to enroll a phase 1, but when the sky’s the limit on combos where there’s already some comfort and clinical experience, the competition for patients and investigators is going to intensify.