Search This Blog


Saturday, December 31, 2016

2016 PCRI Dr Eugene Kwon Provenge

Saturday, December 03, 2016

History of the Narrative

Respect for the truth!

The narrative, as defined by Wikipedia, is "any report of connected events, real or imaginary, presented in a sequence of written or spoken words, and/or still or moving images." Bullshit is defined as, "is mostly a slang profanity term meaning "nonsense", especially as a rebuke in response to communication or actions viewed as deceptive, misleading, disingenuous, unfair or false." Communicating ones scientific research is a narrative. It becomes BS when the communicators are deceptive, misleading and so on. 

OK. Serepta offered up a narrative to the FDA. The FDA told the rest of us that Sereptas narrative was not BS and approved the drug. Some disagreed. We covered that in the last post and time will let us know if the Serepta narrative is BS or not. In fact it seems that Janet Woodcocks approach was to approve the drug to find out. It's not the best way forward and here is why.


I've talked the fateful airplane trip that introduced Lawrence Corey of the Fred Hutchinson Cancer Research Center with David Fallace of the Alaska Permanent Fund. These two got Juno off the ground leaving the details to be filled in by the cargo cult scientists of Seattle. The unemployed tribesmen were gathered and put back into the watch towers with their coconut and stick antenna headsets. Once again they fired up the ceremonies of another biotech pharmaceutical company. The end result is death and destruction. 

It is a very interesting scientific development. Unlike the DNA manipulation that is the narrative of Sereptas Exondys 5, something very powerful is happening with Junos JCAR015. In July of this year the clinical trial known as ROCKET was put on hold after two patients died from a cerebral edema. The deaths were attributed to a protocol change (cargo cult alert!) that added fludarabine (chemotherapy) to the treatment. Juno convinced the FDA that fludarabine caused the deaths and the FDA let them continue on. The cold hard reality cut into that narrative. Since restarting the ROCKET trial 12 patients have been tested. two more patients have died of the same issue, cerebral edema. Take away the emotion of judging the FDA and Juno. What is happening? What would the mechanism of action be with CAR-T therapy directed at CD19 in patients with B Cell Acute Lymphoblastic Leukemia (B-ALL)?

In a Utopian world we would set a team of scientist on the case. What is happening. In the cargo cults we tend to focus on getting the endpoints (the cargo) we want. When we don't get them we try other avenues without understanding the basic science. We pursue the narrative even when BS has presented itself. Brian Orelli of Motley Fool pointed out, "Fortunately, Juno has other CAR-T cancer treatments, JCAR014 and JCAR017, in particular, which appear to be safer, and JCAR015 might still be useful in patients with other blood cancers, such as non-Hodgkin lymphoma. While this issue may be specific to JCAR015 and late-stage ALL, investors in Bellicum Pharmaceuticals, Kite Pharma, and Bluebird should keep in mind that CAR-T is largely uncharted territory that could result in other unanticipated issues with their treatments."

So we keep on keeping on and wait to see who else dies. The patients are in bad condition coming into the trials. This is a part of the bigger problem of treating end of life conditions as diseases we can cure. This is a narrative and it is BS. People being turned into patients for biopharma greed is what we are witnessing. If Juno, Bellicum, Kite and Bluebird can statistically demonstrate longer life (quality of life is not an issue) they can make money. So far Juno has demonstrated the opposite. 

Wednesday, September 21, 2016

Sereptas Cargo Plane Has Landed

“there were serious methodological concerns identified by FDA,” according to the documents.

To measure the drug’s effect on muscle function, the company performed a six-minute walk test on the trial’s participants. The FDA reports there was “no nominally significant difference” between patients taking either the higher dose of eteplirsen, the lower dose or the placebo. The agency also griped about the fact that the company chose to compare the performance of the patients on the six-minute walk test against “historical controls,” or DMD patients who were in different trials in the past. - Forbes

You can read all about the "serious methodological concerns" here.

The simple description of Sereptas plan to demonstrate the efficacy of their product:

1) expression of an altered messenger RNA in muscle (pharmacodynamic)
2) production of dystrophin protein in muscle (pharmacodynamic)
3) improvement or preservation of muscle function (clinical).

Throughout the approval process, critics have expressed concerns with the small population — 12 patients — involved in clinical trials for the rare disease, along with flaws in how the clinical study was designed. These factors make “judgment on science difficult,” Califf (the FDA commissioner) said.

No it doesn't. More data points lead to better understanding of the accuracy of what is being measured. It is simple precision and accuracy. You may leave college with a poor grasp of the math behind statistics but you get the concept. The more data points the clearer the picture gets. Poor precision leads to greater doubts about accuracy. You aren't making a "judgment on science" that is "difficult". You are doing math.

Here is an example of data on the production of dystrophin protein in muscle.

Western blots and immunofluorescence were used to quantitate dystrophin.

Table 2: Applicant’s Quantification of Dystrophin by Western Blot and Immunofluorescence Analyses

Patient Western Blot % of normal Immunofluorescence % positive fibers
A 2.05 18.5
B 1.15 19.1
C 0.38 33.5
D 1.62 24.0
E 0.52 21.5
F 0.98 12.8
G 0 7.1
H 2.47 20.7
I 0.96 28.2
J 0 1.4
L 0.14 4.5

A quick glance at Figure 1: Correlation between Two Methods Used to Quantify Dystrophin in Skeletal Muscle: Patients from Study 201/202 tells you all you need to know. As the FDA put it, "Of note, the correlation between the two independent methods used to quantify dystrophin in muscle samples was weak." They also stated, "As discussed above, we believe that immunofluorescence analysis (percent positive fibers) is not a reliable method to quantify dystrophin content." What other criticisms of Sereptas methodology did the FDA panel mention?

Regarding the first pharmacodynamic goal, to demonstrate expression of an altered mRNA in muscle the FDA states;
Because even a minimal PCR signal is interpreted as “positive,” this biomarker provides little support of efficacy for eteplirsen; it does provide evidence that eteplirsen causes at least some degree of exon 51 skipping, as intended.

Regarding the second pharmacodynamic goal, the demonstrate production of dystrophin protein in muscle the FDA states;
FDA conducted an inspection of the facility where the data reported in the publication were generated. Significant methodological concerns were identified, which cast serious doubt on the reliability of assessments from the first three biopsies.
Thus, the review team does not consider “percent dystrophin‐positive fibers” to be a meaningful way to estimate dystrophin content, and we believe the results reported by the applicant on this measure do not establish that a significant increase in dystrophin occurred in response to eteplirsen treatment
In any case, the level of dystrophin was 0.9% of normal after 3.5 years, such that, in absolute terms, the increase from baseline would be, at most, 0.9%, assuming a “worst case” for untreated patients, i.e., zero dystrophin.

Regarding the clinical goal, improvement or preservation of muscle function, the FDA stated;
Two patients in the 30 mg/kg group became unable to ambulate soon after the study start. The applicant then pooled the six remaining eteplirsen patients and compared them to the four placebo patients, an unplanned post hoc analysis. No nominally significant difference between eteplirsen and placebo was identified in that post hoc analysis.
The applicant conducted a number of additional post hoc analyses, comparing the 6 patients who received eteplirsen in the 24‐week double‐blind phase of Study 201 and could still ambulate at the end of Study 201 (and continued on open‐label eteplirsen in Study 202) to those originally treated with placebo in the double‐blind phase of Study 201, and later switched to open‐label eteplirsen.
The applicant conducted a post hoc comparison of the patients in Study 201/202 to data from the “Italian DMD Registry” and the “Leuven Neuromuscular Reference Center” registry.
The problems of externally‐controlled studies are well recognized.

Dr. Ronald Farkas, who led the FDAs clinical team in the neurology products division, suddenly departed the agency just before the approval of Exondys 51. The arguments for the governments approval of Exondys 51 not come in a highly detailed document such as the one Dr. Farkas and his team presented.
Both FDA camps had “exercised reasonable scientific judgment,” Califf found, adding that it’s “exceedingly rare” to overrule a decision by the director of the Center for Drug Evaluation and Research. Without any additional technical expertise of his own, Califf said, he deferred to CDER Director Dr. Janet Woodcock.
An appeal to the authority of someone elses knowledge!

The jury is still out. Serepta still has to put up or shut up. The FDA has just bought them some time and money. How much money? $300,000 per year. Good grief. With BS artists like these, who needs scientists anymore?

Sunday, September 11, 2016

Risk of Cheating Barometer

Cheating in all areas of life is something that will always exist. We frown upon cheating. We throw people in jail if they cheat others out of their hard earned money. It is in general a bad thing for someone to get something through trickery or fraud. The problem is that we all cheat to some degree. We try to pay as little taxes as possible. We beef up our resumes. We tell little white lies when the need arises. That is why we have skepticism. We need a homeland-security-esque risk of cheating threat level system.

If Hillary Clinton debates Donald Trump, the threat level for cheating should be high for both candidates. They will cheat by exaggerating their accomplishments and lying about their failures. If you like Hillary you might think that Trump is the only one cheating up there on the stage, and vice versa. If you are a fact checker for the debate you operate under a different belief. Your job is to check what they say and compare that to the facts. That is the most honest person in the mix.

If you are a scientist and your job is to create new drugs, you are operating at a high level risk of cheating. The FDA will serve as the fact checker. Your boss is a different story. Your boss is also under a high level risk of cheating. He or she may have a lot of pressure to get the latest antibody drug to show the kind of data the CEO can show the board at the next meeting. Does this mean you are going to cheat? No. But you might end up on a path that will lead to the termination of your project/job. What does an honest person do under these circumstances?

The last post I put on the blog was about race. Black vs white in America is a high risk of cheating topic. "Hands up don't shoot" was the mantra after the Mike Brown incident in Ferguson MO. It is highly probable that that was an inaccurate description of what took place. Facts not supporting the narrative include a discharged weapon prior to the final lethal shots and the strong armed robbery committed by Brown prior to the incident. At his funeral a friend stated the Mr. Brown was out spreading the word of God prior to his death. That narrative would indeed make everyone skeptical about the police departments conclusion that the officer who shot Mike Brown, Darren Wilson, was defending his life. Who needs to shoot and kill a person out spreading the word of God? The problem is that the facts of the gunshot and robbery have been verified. The preaching has not.

Race is a topic where our Risk of Cheating threat level will be high. Was Mike Brown feeling invincible the day he died or was he out saving souls? That information is not proof positive of anything. It is information that only supports a narrative.

Back to BS - biotech science.

Our narrative is simple. We apply basic research as avenues to treating and/or curing disease. Science is the most honest and pure way of thinking. We set aside our biases and stick only to the facts. Narratives involving our motivations (money, fame, altruism...) are not factors in our scientific method. Those who have chosen to question underlying threats to our honesty are not understanding how science works. That is the big picture narrative behind the benevolence of BS - biotech science. We make drugs to help people because we have dedicated our lives to serve our fellow humans. Just don't bring up the history of that narrative.

Quick example: My favorite cargo cult of Seattle WA, Juno. August 31, 2016: Juno CEO Hans Bishop nets $1.28M from 42,673 shares of his stock. Sept. 8, 2016: Juno stock surges on positive clinical trial data.

The stock dives after four deaths in a trial of 129 patients are reported. It goes from $40-something per share to $20-something. The stock creeps back up to around $30 per share. The CEO socks away a cool million in spite of the narrative pointing to a bright future. Threat level HIGH!

Beware investors! Bullshit level high. Store your capital far from the shores of biotech.

Saturday, September 03, 2016

Inside Baseball and why judges should take it easy on black people

This has nothing to do with biotechnology cargo cults. This is about the cargo cult of race relations in the U.S.

Colin Kaepernick has done us a huge favor. He opened the door to let in the non-cargo cult thinkers. For every protest there is a counter protest. We fight to the death to defend even those whose speech offends us. Now they must be willing to hear the other side. Colin Flaherty is the other side of BLM and Colin Kaepernick.

Sunday, May 22, 2016

Should Science Be More Boring?

Price is what you pay. Value is what you get. -Warren Buffet

Try not to become a man of success, but rather try to become a man of value. -Albert Einstein

One of my favorite watchdogs of science is Retraction Watch. They have taken a far better path than I to point out the fly in the ointment of scientific research. In our common way of thinking, scientists are the most honest people in the world. Their discoveries have led to all of our modern comforts, including life saving medicines. As a result the general public has come to think of scientists much the same as they used to think of religious figures. If a scientist publishes a paper on his/her work in a scientific journal it must be Gods honest truth. How else could the scientist have gone through such rigorous judgements to get their work accepted? Retraction Watch is not saying how or why. They simply shine a light on the things that scientists and publishers got wrong. And that is a good thing!

Many people choose to criticize the whole concept of pointing out negatives in a positive world such as scientific work. For example, in the recent Tweet "Science Should Be More Boring" the following responses were posted:

True but in today's marketplace it is not very practical. -Elizabeth Leary

Who's gonna pay the bills with boring articles? -Sheila Shakoor

Agree but that will mean fewer science stories. Unsurprising research is less newsworthy. -Simon McGrath

As much as I agree with publishing negative results, the marketing here stinks. Will Dichtel

Now it behooves me, of course, to tell you what they're missing. But it would be just about as difficult to explain to the South Sea Islanders how they have to arrange things so that they get some wealth in their system.

Aerodynamics, to most people, would be considered boring. The thing is, it is pretty damned important in the wealth of our system versus the wealth of the cargo cult system. The fact that civilized societies have useful things like airplanes is due to boring science. It is not boring however to the people who tend to obsess over understanding how things fly. The real question that follows from the above four comments is what is boring and what kind of non-reproducible science is preferable?

If you send an underling into the laboratory to conduct an experiment on the basic science on the Zeka virus, do you allow for that person to bring you results that they feel will best suit their career ambitions? Or do you expect them to present to you a completely transparent and clearly articulated accounting of what took place in the lab? At what point do we allow the above ideas (on practicality, bill paying, newsworthiness and the marketing of science) take precedence?

To link this concept to a real life biotech situation I must bring up the price of Seattle biotech investment and the value that has been returned. It was the marketing that raised the money to pay the price of funding the companies. It was the value created that led to the vast graveyard of failed companies.

If Sheila Shakoor came to me with an idea for a biotech start-up I would have to wonder, does she have good science as a foundation for her new company or has she been focused on paying the bills? If Simon McGrath has 300 publications on his resume should I be concerned about his focus on newsworthy research? As for Elizabeth and Will, I would ask them to stay in their side of the office and leave the science of the company up to the scientists.

The best comment from this tween was from Stephano Tonzani, "My prof of math methods for physicist once said "for researchers, science is boring 364 days of the year."

I do not find any of this boring. What makes science... science? Reproducibility is one way of truly assessing the value of published work. This idea is also one that meets great resistance among career minded scientists. But if you scroll through the latest postings from Retraction Watch you will find this article from Andrew Gelman.
Whatever the vast majority of retractions are, they’re a tiny fraction of the number of papers that are just wrong — by which I mean they present no good empirical evidence for their claims.
As someone who had to use these papers on a daily basis to conduct "newsworthy" research in the field of biopharma, I am far from bored when I read these words. For thirteen years I struggled with non-reproducible research and an inflexible chain of command that insisted the outcomes be true. In order to put wealth into our system we must change the attitude that subjective valuations on boring versus practical is what matters. Boring science can also be non-reproducible useless information. What matters is not the success of getting published. What matters is the value of the published paper.