Computer Aided Oncology Conference

Every week I go to a few conferences in which cancer cases are discussed.  Oncology, the branch of medicine that treats cancer, is a very active area.  Developments are reported daily.  The rate of clinically important discovery is accelerating.  

I have a good memory, evidenced by my performance in medical school  and  National Boards exams.   But my computer remembers much better than I can.  It never forgets anything. My computer can access the world's medical literature in seconds.  It can translate papers from Chinese.  It can give me the most recently updated recommendations from the NCCN, or the BC Cancer Agency, or the UpToDate  online textbook instantly.  That is why I come to conference armed with my computer linked into the internet.   

 The continuous improvement in computer intellectual power is demonstrated by their relentless ascendancy in high level games.  Chess, Jeopardy and Go are contests in which  complex decision processes are tested and demonstrated.   In 1997, Garry Kasparov lost the world chess champion to a computer,  IBM’s Deep Blue.   It was predictable that the machine would, eventually, win.  IBM's Watson became the greatest of Jeopardy champions. Google’s AlphaGo  is sweeping the Go world.  All of these games, whose masters have extraordinary talents and abilities, are yielding to the machine.  The machine has qualities that no human will ever be able to match... unassisted. 

The machine never forgets. The machine can search its memory, most of the world's knowledge, with lightning rapidity.  It can carry out algorithms at near light speed. The puny, slow moving human memory is no match. 

But humans have qualities that the machine cannot (yet) replicate.  The human has sensitivities, shared in a multifaceted and spectral way, with other humans.  There are ways in which we understand each other that a silicon processor cannot.  It could, possibly, be programmed to mimic the human responses to emotion, but it cannot feel empathy. 

When I bring my computer to conference, I add to the machine.  I tell it what to look up.  I define the search parameters.  Having done this for more than 25 years, I can do it rapidly and well.  It is an art that incorporates a level of speculation about how biological processes might be related.  It also involves a (reasonable) expectation that someone else has had the same question and it has been written about.  

I then decide which articles are truly relevant and estimate the validity of the information.  Even a human brain can do all this in less than two minutes.  I think it works out well.  That is why I continue to do it and encourage others to try it, and improve upon it. 

 The machine is spectacular in the things it can do.  The doctor adds human qualities and direction to that instantly accessible, vast, and reliable knowledge base.  No person can beat the machine in its tasks, but the combination of the human and the machine remains (for the foreseeable future) the best solution, better than the machine alone. 

The presentation of the patient's case at conference and sharing of experiences and opinion  is critically important.  It forces an organized and clear presentation. All the data (pathology, imaging, etc.) is open for inspection and scrutiny. Sometimes, carefully considered human analysis  prevails over all the data.  

  Better data can help create better opinions.

  Let’s improve. 

Tortoise and Hare

Tortoise and Hare

The profound truth of Aesop’s favorable can be seen in oncology.

We all remember the story:  the tortoise and the hare agreed to a  race.  There is no doubt about who is the faster.  But as a consequence of his early lead, the hare falls asleep and loses the race.

The same phenomenon happens in the treatment of cancer.  There are often breakthroughs, instances of rapid progress in the understanding and treatment of various malignancies.  Industries develop around these advances.  The advances become standard practice.  But the adoption of these new treatments can stand in the way of developing even better treatments.

Checkpoint inhibitor therapy, medicines that block the immune system turn off valve, allowing a more effective immune-driven attack on the cancer, was first developed and approved in melanoma.  There was really no effective treatment for advanced melanoma 10 years ago.  It was an l oncologic tortoise.  Checkpoint inhibitor therapy is not specific to any particular cancer.  There is no clear relationship to melanoma.  But there was so little to offer melanoma that trials could quickly proceed followed by prompt approval of these medicines.

Now checkpoint inhibitors are being studied in a variety of cancers.  They are already approved for squamous cell lung cancer and commonly used for metastatic renal cancer.  These are two other cancers  for which therapeutic options are limited.  But this type of treatment has no clear relationship to the cancer histology.

In cancers that have an array of somewhat more effective treatments, the introduction of checkpoint inhibitor therapy has been much slower.  The tortoise that has become the hare,  cannot enter the race  because of previous rapid advances.

The use of next generation sequencing for the classification of hematologic malignancies is also a case of limited advances blocking the introduction of deeper techniques.  Prior to the introduction of relatively rapid DNA sequencing techniques, hematologic malignancies ( leukemias and lymphomas) were the most extensively molecularly  characterized malignancies, Chromosomal rearrangements defined diseases and directed therapy.  Clonality, a fundamental property of malignancy,  was easily and standardly determined.  Disease could be followed on a molecular level.

However, heterogeneity remained.  Rarely, patients who had a very high probability of excellent response could be identified on the basis of these chromosomal, FISH and limited PCR tests.  But the details remained mysterious and thus the response to treatment remained unpredictable

The introduction of next generation sequencing for the characterization of hematologic malignancies has been slower than from many solid tumors.  The advances made with older molecular techniques have blocked the introduction of the more detailed and extensive analysis now available.  On a molecular level, hematologic malignancies have been asleep

I think there is value in this awareness.  The extreme sub-specialization In oncology has limited the diffusion of advances from one field to another.  The economic arrangement of the healthcare system favors maintaining diagnosis and treatment on a plateau of adequacy It should be helping advanced medicine to increasing excellence.

Sleepers awake

The Box

Lymphoma Rounds: In the box

Should doctors think in  the box?  Sure, if the box works, If the therapy is effective, if the criteria for diagnosis define an entity for which the treatment is successful, barring a complicating factor, then one should follow the guidelines, the textbook.  .  One should be sure about identifying the problem and choose the best solution and do it.There are situations in which  this scenario applies: many infections, most Hodgkin's Lymphomas, many testicular cancers. Once the diagnosis is secured, the exceptions ruled out, and the patient tolerance assured, following the formula will yield the desired outcome.

I went to lymphoma rounds.  The first case was a case of lymphomatoid granulomatosis. This is an unusual diagnosis.  The case was more unusual because of CNS signs and symptoms.  The case was presented by a fellow and discussed by a neuro-oncologist. That meant that the neurological aspects o the case were made  primary. 

The relationship to Epstein-Barr virus in this lymphoma was clearly stated and recognized. We had all learned the pathophysiology of EBV disease: cells are transformed, changed by EBV into cells that share many qualities with malignant cancer cells.  They are immortal, The program for cell death is turned off..They reproduce out of control.

The acute disease, mononucleosis, usually comes to a good conclusion withe the patient's T cell based immunity victorious.  But there are EBV driven lymphomas that are not self limited, lymphomas that  do not improve without strong, chemotherapeutic intervention.  These include various lymphomas in the immunocomprimised. Patients with  (uncontrolled) HIV disease can get  aggressive lymphomas, including CNS lymphomas,  from EBV.  One of the most aggressive of all lymphomas, Burkitts lymphoma, in its classical form, is an EBV driven disease.  Patients who have had organ transplants and take immunosuppression  can get less aggressive  lymphomas like PTLD, a lymphoma that often responds to changes in immunosuppresion.  EBV disease often has an immunological basis

The case was discussed in terms of published treatment outcomes, the standard therapy was CHOP, all purpose lymphoma therapy.  Some mention was made of interferon, which seemed to succeed in some cases, but was not used in this case. 

I asked if the this entity could be divided into different groups based upon the immunologic status of the patient,  and the nature of the cells identified in the biopsies.  After a long silence, I said that I did not expect and answer to the question.  (This patient's immunologic status had not been investigated.) 

The true meaning of the question involved the approach to such diseases.    We live in an era of instant access to millions of research findings.  Almost anything you can think of has been written about,  What is written is not necessarily good, and may even be misleading, but should not be ignored. 

When a disease is outside the of the set of the clearly curable with standard approaches, the consideration should include an approach that incorporates our beliefs about pathophysiology.  Such an approach can lead to other pathways for diagnosis, treatment and prognosis. 

I think that there is always an immunological aspect to EBV disease. The immunologic aspect has implications for HIV related lymphoma, encouraging anti retrovirals and the reconstitution of CD4 mediated immunity.  We exploit it in PTLD by adjusting immunosuppression ( sometimes  along with Rituxan)  Even when there is central nervous system involvement, simply reducing immunosuppression can be sufficient for remission in PTLD.

I do not completely trust our (current) understanding of pathophysiology.  It will probably change with the acquisition of further knowledge, but I am not prepared to take it out of the picture, 

Hypotheses about the nature of the patient's disease can be lifesaving.  They can also be terribly misleading.  Interpretation requires great care and open discussion 

The Martian: How much is a person worth?

The Martian: How much is a person worth?

Recently,  I saw The Martian, a movie about a man accidentally abandoned on Mars.  The movie is about the struggle for survival, the marshaling of forces to allow the survival of that one man,  the sacrifice of  compatriots and the politics of rescue .  

Cancer patients, and the people who care for them, can feel like they are abandoned on Mars. The feelings elicited by this movie are similar to those we, who care for cancer patient patients feel. When do we call the situation hopeless?  When do we give up?  How much can we put into the effort for one patient?  How much can we spend?

In the movie there is no limit. Billions of dollars are spent,  scores  of people work without rest, people give up years, in the prime of their lives, to attempt to rescue a single man.  In our real, medical world the money, the time, the energy are all limited,  The resources are shared by thousands of patients. This places every part of the medical system in the position of distributing a limited, precious resource.  The doctor must balance the chance of benefiting the patient against the cost to the system, which could mean denying another patient an equal or better chance.  Doctors differ in their approach to this problem.

How can we do any less than our best?   Our efforts are not like those in the movie.  They are not as good as they should be.. The basis for saving the Martian was adoption of a nonstandard strategy, a strategy that would work, in theory, but was not a usual approach.  A methodology that involved unanticipated expense and sacrifice.

Currently, the pressure to follow standard procedures is almost overwhelming.  Deviation from such standards  risks the label of malpractice.  Obtaining insurance coverage for a treatment that is not recommended in guidelines, or for a problem that deviates from the FDA  approval parameters is a Herculean task - and getting harder.  .

Knowledge and Resources are always limited. The Martian was rescued, he beat  the odds. It is very expensive and difficult to take on the odds... sometimes it works.

A better way to treat cancer patients

There is a better way to treat cancer patients

I am a practitioner of oncology.  When I am confronted with a treatment decision, I attempt to bring the best data to bear on the subject. I avail myself of resources: online text books, PubMed, Web of Science, Clinicaltrials.gov, Eviti advisor. I try to gather and compare  the options  I very much want to see which option will help my patient best (longest lasting, least toxic, most likely to be effective).

But the medical literature does not allow me access to detailed information.  It tells me the outcomes of large groups of patients, diverse in the details of their disease, expressed as probabilities over the whole group and sometimes subgroups, defined by the authors ,that are usually not helpful.  I think that the details may direct the treatment; but details are not available.

Instead of selecting patients who are suitable for clinical trials, we should be constantly improving our treatment, we should be using an iterative correction model,

Patients should be encouraged to consent to have  information that could be related to their disease and its treatment, stored and available for real-time analysis. The data should include a great depth of detail,   Every time a patient comes for treatment we should check the current,  updated  outcome  for that treatment in patients that are most similar to the one we propose to treat. That information should correct our initial proposal.  Once a final decision is made, that patients course becomes part of the growing, searchable and expandable database.  That patients course becomes part of the decision for the  next patient.

There are problems: privacy, accuracy, validity. These problems are the subject of current research [1]

This is a call  for revolution in how clinical data is reported and accessed. I want a  database of raw data, available for appropriate analysis along parameters relevant to  the patient at hand.  I would like to be able to collect information on patient's who are similar to the patient under my care.   The similarities would be defined by the current and emerging understanding of the disease and its treatment. 

The technology exists to start such a program.  As more as learned, the program will improve  Internet service providers already follow this model and the effectiveness of that effort is reflected in the fact that people pay for it and profit from it.  That same process should be used to help save lives.

Paradigm Shift in Genomics

The presumption  had been that the  genomic defects in cancer, revealed by sequencing data, would identify targets for treatment.  By virtue of the specificity of these targets, treatment would be more effective and less toxic. In a few situations, this has happened. The EGFR , ALK  ROS genes of lung cancer, the B RAF of melanoma, HEr2 in breast and esophageal, abl in chronic myelogenous leukemia are examples of the success of this approach ( N.B. all antedate widespread next generation sequencing.)  The estrogen receptor of breast cancers  and the androgen receptor of prostate cancers are therapy targets that are not correlated with sequencing data.

I do not want to be overly cynical about the value of sequencing.  The the technique, as clinical art and science, has not matured.  Development often  produces surprises.  When the surprise  leads to  a broader model, the  original idea needs to be placed in the new perspective.

Recent discoveries have broadened the meaning of  NG sequenced data.  Rivzi, et al  have shown that response to immunomodulatory, PD-1, directed therapy correlates with overall mutational burden. A tumor in which a large number of mutations is identified by NG sequencing tends to respond more favorably to an agent ( monclonal antibody) that blocks an "off switch" for immune reaction ( PD-1).   It is the mutable tendency, not the actual mutation that determines responsiveness to this (class of) agent(s).  The observations of Le, et al, that mismatch repair, a phenotype that would be expected to increase the mutation rate, signals sensitivity of cancers to  PD-1 blockade seems to be a complementary  finding.

These findings imply that the mechanism underlying the cancer, tolerance of mutation, rather than the mutations themselves,  can be a target for treatment.  This particular mechanism also intimates that the immunologic treatment will eventually fail since mutations will continue to occur, and be tolerated.  in the malignant cell population.

The utility of PARP inhibitors in BRCA and PALB2 mutated tumors is a corollary to these ideas.  The PARP inhibitors target the mechanism of mutational generation in these selected cases.

Understanding optimizes hope

The practical use of clinical trials

Clinical trials are very expensive.  Current funding from the NIH for clinical trials is : $7,327,474,432 (7 billion dollars) I do not think we are getting our money's worth.

When I treat a patient, I want to know the best treatment for my patient, not the best treatment for a population of patients who have (what the trialists have called) the same disease, and who fit the criteria for the trial ( i.e. no other major problems). I need to optimize the outcome of the patient as I see her.  But the reporting of clinical trials obscures the information I need to make the best decision. The details are sacrificed to an illusion of statistical accuracy.

I have a patient who was treated for "multple myeloma."  He has the characteristic abnormal antibody product ( monoclonal spike) resulting from a cancer of plasma cells( the immune cells that produce antibodies),  Chromosome analysis   demonstrates a translocation (4;14)  This chromosomal abnormality identifies a disease that has an inferior outcome with (old) standard therapy.   There is evidence that newer treatment  ( boretzimib)  produces a better outcome.  The patient had standard therapy including a stem cell   (autologous) transplant.  What should the mainteneane therapy be?

The studes demonstrate an advatage for mainetenace therapy,  But the results are expressed without regard to the chromosomal abnormality. At best, this chromosomal abnormality is lumped with others that define "risk"  But risk is a function of treatment!.  To make maters worse, the risk classification of his chromosomal abnormality varies among authors!  Some call it high, other intermediate.

If the raw  data were available, if the clinician could aggregate and analize the raw data, and then submit the data on his patient into the pool, we would have afar more efficient and productive  process.  Instead of basing decisions upon random data, we would base decisions upon outcomes that were also based upon constant correction,  The search for optimization would be  organic, ever improving.. Conceptually this approaches grows out of Bayes ( using the pretest assumption)  and Hayek ( the wisdom of the market).   This is an optimization algorithm,  Why shouldn't each successive treatment attempt be corrected for the past treatment attempts?