Weeks 174-176: Poison for Christmas, delirium for New Year

Written by Lars Haakon Soraas

Weeks 174, 175 and 176 in the war on lung cancer have passed. The lack of updates has not been because of Christmas and New Year. It has been because we have been, and still are, in a frenzied fight for survival. There is so much to write and share and I don’t quite know how to structure this blog post. I will try to do it more or less chronologically. Here goes:

  • As you may recall, in our last post, Dyanne in December had surgery done to remove a tumor above her right collar bone. The main purpose of this was to get tissue that can be used for analysis. Part of the tissue was sent for drug sensitivity testing. In short this means testing a range of drugs on the cancer cells and then seeing which drugs, or drug combinations, are most effective at killing the cancer cells.  On Monday and Tuesday 17th and 18th December we received results from this testing. I will write more about the results below, but of immediate concern for us then was the fact that the chemo combination that Dyanne was scheduled to receive on Wednesday 19th December did not do very well in the test. One of the drugs, carboplatin, seemed to do quite well. But the other chemo drug, pemetrexed, seemed like it would have virtually no effect at all. A couple of other chemo drugs, etoposide and irinotecan, seemed like they could be promising. Common for these is that they are typically used in small cell lung cancer (SCLC), not the non-small cell lung cancer (NSCLC) that we have been battling over the past years. This, combined with a couple of other observations we had made of the tumor (the fact that it harboured both a TP53 mutation and RB1 loss, and the location and speed of its growth) made us strongly suspect that what we were experiencing was a so called NSCLC to SCLC transformation. This means that the tumor basically transforms (in part or completely) into a different type of tumor. This happens in maybe 5-10% of EGFR mutant lung cancers and would thus, a priori, not be that unlikely. As chemo sucks to begin with, doing ineffective chemo is one of the last things you want to do. Hence, after some intense hours of deliberations, we concluded it would be best to change the chemo from carboplatin and pemetrexed to carboplatin and etoposide. We called our oncologist late on Tuesday and asked if this could be possible. Thankfully she was in agreement and managed to get the change organized for the chemo infusion that was scheduled on Wednesday morning.
  • Dyanne then got the chemo infusion on Wednesday 19th December. It was quite uneventful, even if she got an allergic reaction of some sort, needed to get some antihistamines and 4 mg of dexamethasone. She also took etoposide as tablets on the following two days.
  • As has been mentioned in previous posts, we measure a number of tumor markers regularly every week. These tumor markers are basically simple blood tests that can help you track if the cancer is growing or not. Two of the markers we are tracking, CEA and NSE, had gone wild in the past months and exhibited exponential growth. On the day before Dyanne started chemo, CEA was 99 (normal level is less than 5) and NSE was 75 (normal level is less than 12). We then measured these again on Friday 21st December, only two days after Dyanne did the chemo infusion. We then got the first signs that chemo was working: CEA had dropped from 99 to 93. NSE had dropped from 75 to 46. These markers have continued to fall and on the latest reading we have of these markers, which is from 31st December, CEA was 47 and NSE was 11.
  • As is widely known, chemo sucks. There is thus no point going into great detail, but Dyanne has had nausea (not too bad), fatigue (quite bad), neutropenia (not too bad – at least not yet), a bit of anemia and now, in the past few days, quite a bit of hair loss.
  • In Norwegian the word used for “chemotherapy” is “cellegift”. The direct translation of this to English is “cell poison”. So for Christmas Dyanne got poison. The good news is that the poison at least seems to have some effect against the cancer.
  • As mentioned in the past blog post, the December MRI scan of the brain showed that there were two metastasis at the base of the skull. These were pressing on a nerve and causing numbness in her right chin and the right part of the tongue and gum. We wanted to get these radiated quickly and luckily our hospital here in Oslo managed to get this organized quite quickly. On 21st December Dyanne got the first of radiation treatment of these two lesions. The second dose was on 24th December and the last dose was 27th December. Each time the dose she got was 6 Gy. So in addition to poison, Dyanne also got radiation for Christmas.
  • Then, on 23rd December we got the report from the pathologist who looked at the tumor that was removed surgically earlier in December. He made a number of interesting observations. First of all he found that Dyanne’s tumor was not any longer NSCLC. Contrary to what we had believed, however, it was not SCLC either. It turned out it was large cell neuroendocrine carcinoma (LCNEC). Both SCLC and LCNEC are neuroendocrine tumors and typically respond to the same type of chemotherapies. Our decision to switch from pemetrexed to etoposide then seemed to be sensible. The pathologist also confirmed that the tumor was aggressive. This is something we, of course, already kind of knew. It was nevertheless chilling to read the pathologist’s finding that the tumor “has many features suggestive of very aggressive behavior”. A marker of aggressiveness that is commonly used, Ki67, was 100% (scale goes from 0-100%). In case anyone is interested, the pathology report is pasted in as a picture at the bottom of this post.
  • How bad is LCNEC? Turns out it is pretty similar to SCLC. Median survival for those diagnosed with metastatic LCNEC or SCLC is a bit less than a year. A paper that came out just a few weeks ago showed that median survival of EGFR mutant lung cancer patients who experienced SCLC transformation was 11 months (from the time of transformation). I paste in a chart from the article at the bottom of this post. The chart shows how long patients that were included in the study survived after their tumors transformed to SCLC. The song “Last Christmas” playing on the radio suddenly got a different meaning…
  • We started reading about LCNEC. The “good” news is that it is such a rare condition that there is really not much written about it. EGFR mutant lung cancer transforming to LCNEC is only reported a handful of times before. The perhaps most interesting paper we read was this one: Integrative genomic profiling of large-cell neuroendocrine carcinomas reveals distinct subtypes of high-grade neuroendocrine lung tumors. It classifies LCNEC into two subtypes (“type I” and “type II”). Based on testing we had already done on Dyanne’s tumor tissue, could classify Dyanne’s tumor as “type II”. This type of tumor may be responsive to immunotherapy, as suggested in the paper, and as is also suggested by this paper and this abstract. One reason to worry that immunotherapy may not work in Dyanne’s case is that her tumor only has few mutations, while LCNEC typically tend to have many mutations (which many believe increase the chance of response to immunotherapy). Nevertheless, based on these findings we are considering doing more immunotherapy.
  • Then back to the results from the drug sensitivity testing. I will not go into great detail on this, as it could deserve its own blog post (or, in fact, several blog posts…). I will mention this, however: three of the drugs that came out best in the testing, apart from the chemo drugs, were disulfiram, auranofin and panobinostat. I may comment on auranofin and panobinostat in future posts, and will here commment only on disulfiram. Disulfiram, sold under the brand name Antabus, is a drug used to wean alcoholics of alcohol. If you take the drug and drink, you get really sick. And knowing this you basically won’t drink. There is lots of research showing disulfiram has anti-cancer effects, including in lung cancer. This editorial gives a nice overview. Disulfiram also happens to be cheap and low on side effects. So what do you do when a cheap, virtually non-toxic drug with lots of promising anti-cancer data comes out close to the top in the drug sensitivity test you have done? Well, you start taking it.
  • So Dyanne started this just before Christmas. We read a lot about it before, and also after she started taking it. It is not difficult finding articles suggesting it may be a good thing to try out if you have terminal cancer. One thing that is curious, however, is this: Dyanne is allergic to alcohol. She has two copies of a gene variant (ALDH2 E487K) that basically makes her body very bad at metabolizing alcohol. If she drinks, she gets sick pretty much immediately. Among East Asians this is common: around 40% have this gene variant and the the fact that they get red in their face has it’s own Wikipedia entry: https://en.wikipedia.org/wiki/Alcohol_flush_reaction. Dyanne has two copies of this gene variant, making her particularly allergic to alcohol. So the question was then this: what happens when someone who is really allergic to alcohol takes a drug that is meant to wean people off alcohol? We hoped this would spell good news. Perhaps, for instance, that her cancer cells would be particularly sensitive to this drug and all go off and die.
  • Of course, if something seems too good to be true (a cheap drug with no side effects curing your cancer), then chances are… that it is not true. And we learned this the hard way. On Thursday this week (3rd January) Dyanne started being somewhat confused. Simple tasks (e.g. checking bank account) suddenly seemed challenging. It got a bit better during the day, but at night it was clear that something was very wrong. She was talking incoherently, repeating short phrases, not being able to answer simple questions and even struggling to state how old she was. We called our cancer hospital and they said it was most likely a side effect of the radiation she had gotten to the base of the skull, that it was normal, and would go away. On Friday morning she was not any better and we had a friend (who is a psychiatrist) come over and take a look at Dyanne. She confirmed what we had figured out ourselves: Dyanne had very clear signs of delirium. She told us this needed to be taken seriously and was rather urgent. We called Dyanne’s GP. He was, as always, amazing and gave her an appointment half an hour later. He then quickly diagnosed Dyanne with delirium and had her admitted to a local hospital here in Oslo.
  • The most important when it comes to diagnosing and treating delirium is to find out what is causing it. We started slowly suspecting that disulfiram was to blame. The reason why disulfiram would be the cause is not really clear, but it was a hunch and there are also some suggestions in the scientific literature that could lead one to think it was the cause of our ills (I would like to thank two fellow patients/caregivers, Anne Sofie and Jo, who helped researching this). Further strengthening our suspicion was the fact that the hospital did every test in the book to find out the cause and could not find any plausible source of the delirium. And when I spoke with a US doctor/researcher who has experience with off-label use of disulfiram among cancer patients, and he mentioned he was aware of two other cases where disulfiram was thought to have caused delirium, we concluded that disulfiram was most likely the cause. It was thus decided to stop disulfiram.
  • As a kind of funny, if it wasn’t so serious, side note: that same night the hospital Dyanne was admitted to tried to give her more disulfiram! However, the nurse had been a bit too quick when reading the name of the drug and interpreted it to be a drug with a somewhat similar name. They thus offered Dyanne not disulfiram, but a drug called diflucan. They meant to give her a drug she was absolutely not supposed to have, bungled it, and gave her another drug she was not supposed to get. Dyanne, of course, being in a state of delirium, would never have found out any of this. One of the things I regard as my job in this ordeal is to try to ensure the Norwegian hospital system doesn’t kill Dyanne out of ineptitude, as tragically happened to 6 year old Djabrail. Luckily I was there and bothered to ask what they were up to. And luckily I kept asking questions when they gave non-reassuring answers. I am sure many hospital staff regard me as a pain in the @$$, and all I can say is that I will continue to be one until they get their systems in order (which I don’t expect will happen in my lifetime).
  • Anyway, yesterday morning Dyanne was a lot better. Still with clear signs of being a bit “tipsy”, but now being able to talk coherently. She was signed out of the hospital and we went back home. Today she is even better than yesterday, but as she still has quite a bit of disulfiram in her body, she is not completely out of it yet. It may take a few more days, but it is clearly going in the right direction.

Now, before I round off this way too long blog post, I would like to mention three more things.

First of all: if you are an East Asian and you get red in the face when you drink alcohol, or if you got red in the face after drinking alcohol in the first year you drank, then you quite likely have one or two copies of the gene variant that Dyanne has (i.e. ALDH2 E487K, aka ALDH2*2). And if you have this gene variant and drink, then you have a significantly increased risk of esophageal and also head and neck cancer. And there are also data showing it may increases your risk of lung cancer. Very few seem to know about this, even if it, scientifically speaking, is well established. I have found a great review of all of this and highly recommend every East Asian to read this paper: A Personalized Medicine Approach for Asian Americans with the Aldehyde Dehydrogenase 2*2 Variant.

Secondly, I would like to highlight what I believe is an underutilized property of many cancer treatments. The property is this: if they work, they often work very fast. This is illustrated by the following two points:

  • Two days after Dyanne started chemo, her tumor markers were down significantly.
  • Many patients who respond to cancer treatments report marked improvements in symptoms after 1-3 days.

So if the treatment works, it is often possible to detect this very quickly. Not 2 months after the treatment is initiated. Not even 2 weeks. Often it can be seen after 2 days. Why do I say that this is an underutilized property? Well, because this means that patients could be able to cycle through a large number of treatments in order to find out if anything works for them. Cancer is heterogeneous. What works for one patient with one type of cancer may not work for the next patient, even if he/she has the exact same diagnosis. The way modern cancer treatment works, however, is to start a treatment and then only evaluate how things are going after 2 months. For a cancer patient with a median survival of 10 months, this means the maximum number of treatments that can be tried is around 5. And, due to bureaucracy and an antiquated modus operandi, most cancer patients only try out a couple of different treatments before they die.

In a smart system a cancer patient should be able to try out a new treatment every week until a treatment that works is found. This means they could probably try out one order of magnitude more treatments than what they are doing today. If they actually did this, the chances they find a treatment that works would become much higher. Furthermore, such an approach would significantly increase the innovation of the whole system and it would thus also significantly reduce the time humanity would need to find the cure for this terrible disease. Anyway, we won’t be able to fix this, but I think this nicely shows how bad a system humanity has equipped itself with in the war on cancer. On another planet, surely the mice are laughing.

Finally, we earlier this week learned that a fellow EGFR mutant lung cancer patient, Stephanie Engel, has passed. She was way too young and leaves behind a husband, two children and many others. She was part of our EGFR email group and put up a tremendous fight against this nasty disease. She inspired us and many others. She will be missed.

With that, Merry Christmas and a Happy New Year. Or at least Christmas and New Year.

Kaplan-Meier overall survival (OS) curve for EGFR mutant NSCLC patients who has had their tumor transform into small cell lung cancer (SCLC)

Pathology findings

Recovering from delirium

Follow this blog

Get every new post delivered right to your inbox.