Patient Choices

Treatment choices are often difficult to make.  Here is one approach.

Although I am not a "medical freedom" advocate, my rare cancer has no effective standard of care, so I need SOME kind of approach to make my treatment choices.

I outline how I do it below.

Clinical Trials

After my kidney was removed, I asked my doctor if I ought to do anything else.  He said that I might want to consider an adjuvant clinical trial called EVEREST.  The idea behind EVEREST was to put patients on Everolimus for a while to kill any wandering cancer cells that the surgeon missed.  A Normal person would get a second opinion before making this decision.  After all, Cleveland clinic states: 67% of second opinions recommend a diagnosis or treatment plan change.  I learned more by being abnormal. In particular, I asked a “tumor board” of 13 people to answer the question “Ought Bill Paseman enroll in the EVEREST Clinical Trial to treat his rare kidney cancer?”

• 3 said “Yes” (Turns out that the first three people I asked said “Yes”. )
• 5 said “No”
• 5 said “Bill needs to decide”.

Two of the “no”s were pretty convincing.

Here’s the first “no”: “Kidney cancer has had no successful treatment (e.g. radiation, chemotherapy, immunosuppressants).  So patients with the affliction usually just wait, hope, and live out their lives.  If you participate in the trial, you would certainly further scientific knowledge, but chances of success are low.”

Turns out that “Chance of success” in an oncologic clinical trial is 3.4%.

The second “no” from Nizar Tannir was "I do not recommend any adjuvant trial w/ mTOR inhibitors or VEGF targeted agents for papillary RCC.  There will be trials w/ immune checkpoint agents in the near future but not soon enough to enroll on."

What informed his opinion? Turns out that he tried treating my type of cancer using Everolimus in a (non-adjuvant) clinical trial (ESPN), and it didn’t even slow it down.  So I picked “No”.  Note that if I had just gone with the first three people I asked, I would likely have done the trial.  I have been lucky.  So far, the kidney cancer has not recurred.

Alternative Radiation Treatments

Still have that pesky Brain tumor dead center in my head. I let it grow until 2022.  At that point, it was so big that if I let it grow further, I’d have to have surgery. The other option was radiation.  Either Proton, which stops dead in the tumor, or photon (gamma radiation) which goes all the way through. So I created another “tumor board” of 13 people with the question “Ought Bill Paseman do proton or photon?” It was kind of like visiting an Audi or a BMW car dealership and asking what the best model car was. Hospitals only had proton or photon, and they always recommended whatever they had. However one hospital had both and recommended photon. Another (proton) guy said “You’re asking the wrong question. You don’t ask a garage what wrench they use, you ask what their success rate is at fixing cars.” So I went with UCSF (which uses photon).

Treatment for recurrence

Now, if the kidney cancer comes back, I still need medication. And that is a problem because my disease is rare and existing kidney cancer medications (E.g. Everolimus) have been shown not to work. What to do? Pete Kane and I held a “hackathon” consisting of 17 “tumor boards” of 4-7 people. Each “tumor board” was given my DNA and asked to find “genes of interest”, and an associated treatment.

• Sidebar
  • DNA Genes are transcribed into RNA and the amount of transcription varies by organ.
  • So although the kidney and brain (for example) have the same DNA, the brain transcribes more brain RNA and the kidney more kidney RNA.
  • Similarly for Tumors.  So I used a service to measure RNA expression levels from both my normal kidney and my kidney tumor.

I then used the difference between my Kidney-Normal and Kidney-Tumor expression levels to judge the results (below). Here the bar height is the sum of the expression levels of each of the Team’s “Gene of Interest”. And the graph itself shows the value of the opinion of each member of the game (which itself can be viewed as a tumor board). You can read more about the hackathon here.

Long Tails/Power Law Distributions

To Summarize, 

• 1/13 people gave a credible answer in the “Clinical Trial” test
• 2/13 people gave a credible answer in the “Radiation” test
• 1/17 people gave a credible answer in the “Hackathon” test

I’ve seen these ratios before in Finance.  Two Examples:

• Example 1: I formed a portfolio for my kids in 1997.  I did not touch it.  Here is how it has grown.
  • INTC 47%
  • CSCO 103%
  • PYPL 128%
  • EBAY 209%
  • MSFT 2105%
• Example 2:
  • Kleiner IX, a limited partnership formed 1999, had 38 investments: 37 were OK, #38 was named GOOGLE. 

These two examples illustrate what Peter Thiel said in his book, “Zero to One”: “The biggest secret in venture capital is that the best investment in a successful fund equals or outperforms the entire rest of the fund combined.”

My Question

In a Power-law distribution, the largest entity is typically bigger, more valuable, or more powerful than all others combined. Financial portfolios have a power law distribution. Do “Second Opinions” also have a Power Law Distribution? If so, I need something besides mean/variance to describe them. Here is one idea:

• Jeff Bezos Test: Suppose you have a party for Amazon Employees.  You measure the party's average net worth as guests arrive.  When Jeff Bezos arrives, the average is broken because he is an extreme outlier.  So if you are looking for really good advice, the rule is:  "if answers you've gotten so far fit a normal distribution, then keep looking". One thing that REALLY sucks about that observation is that you are ALWAYS likely to find something better. But only if you wait long enough. And of course, once you find it, there is the temptation to wait for something bigger.