Do you want the best data or patient safety?

jenrio

Very interesting article on the trade-off:

http://www.forbes.com/sites/matthewherper/2012/06/04/new-cancer-data-shine-spotlight-on-the-secret-committees-that-make-medicines-toughest-decisions/

 

I'm naturally on the side of the faster, cheaper processes leads to faster/cheaper drugs, and ultimately more effective drugs.   The cost in delaying the process lies in the patients' lives endangered, more difficult recruiting of patients into randomized clinical trials, and in the ultimately higher cost of drugs.    

 

Note how almost all the drugs that had early terminated drug trial end up being highly effective and lifesaving.    If a couple of drugs later turned out to be less effective than originally thought, it's not a big deal.    The standard for treatment for end-stage cancer with no other options should be different than the standard for treating something less deadly.  

cp418

I want both - proven efficacy and safety that are BETTER than the current treatment - - not more "me too" drugs just to make money.  I do not think a clinical trial should be stopped based upon preliminary results. Reporting postive findings is fine and bad findings necessary in case a trial needs to be stopped for patient safety - toxicity.  Instead continue to monitor the patients as was intended in the study design - - they deserve to be monitored for the sake of safety. If results look so promising then make the drug available to more patients - - with restrictions and monitoring required because the study was INCOMPLETE and not enough data available for full evaluation for efficacy and longer term safety. Once those drugs are approved and then later patients have problems then you have to deal with lawyers to get the attetnion of the drug companies to pull dangerous drugs.  Too mnay news articles of drug companies ignoring warnings and complaints by patients for FDA having to step in and police them.  Time means lives are at stake --- drug companies are about money in the end and they need to be held to specific guidelines regarding the welfare of patients. 

jenrio

What if you can't have both?   That's why it's called a "trade-off".

I think this article does not cover all the costs of a more lengthy process that tries to ensure data integrity. Faster, cheaper processes leads to faster/cheaper drugs, and ultimately more effective drugs. The cost in delaying the process (more data integrity) lies in

1.  the patients' lives endangered,

2. more difficult recruiting of patients into randomized clinical trials.  Patients who put their lives on the line to go through the blinded trials.  When they hear the preliminary data that's highly promising but may be just a little short, they need to make the choice of dropping out of clinical trials if they are in the control arm vs staying in the control arm, and die to make the OS (overall survival) data true.   That is difficult choice for patients and physicians, and difficult for any future double blind clinical tials to recruit.

The difference between "PFS" vs "OS" as end-point for efficacy is ultimately, somebody has to die as a result of being on the control arm, hence making the OS difference significant.   

3. in the ultimately higher cost of drugs.

Ideally, even PFS as an end-point should become obsolete as end-point for efficacy.

http://www.nature.com/nature/journal/v485/n7400_supp/full/485S58a.html

I wrote a review for this article in my blog. 

Different clinical trials should be conducted, with different end-point. The ultimate goal is to make the turn-around-time fast. So scientists can try a drug, wait a few days/weeks for results; tweak it to improve it, wait a few days/weeks for results; repeat above.

Currently the scientists must wait years and years for results every single human related iteration of drug design. No wonder effective treatments are few and far in between and very very expensive.

Imagine one day there is a CTC test similar to blood glucose test, where patients can poke their fingers in morning/lunch/dinner time and get 3 data points. Patients/physicians can see in real time how CTC is changing with new treatment/exercise regimen, and this data can be used to forecast patients' real response to medication and used in clinical trials, which takes a few months, even weeks to run, instead of years and years.   Patients need not put their lives on the line to join a randomized clinical trial.  And 100% advanced breast cancer patients can join a clinical trial that generate a valid data point without having to make a commitment for years till death (for the OS benefit).    Then we can say truly, that MBC is a chronic disease, like diabetes. And its cure would not be far off.

cp418

I do not believe "Faster, cheaper processes leads to faster/cheaper drugs, and ultimately more effective drugs".  I believe well designed studies with the best drug candidate with accurate data collection leads to better evaluation of safety and efficacy.  Why approve or take a drug if it may end up worse than the current treatments available?  IMO more compassion trials should be opened up with different guidelines and as such with different study design and enrollment and open treatment criteria. 

Pharma companies in recent years now use electronic data collection.  No more paper with patient study books and manual data entry.  Yes, there are entire clinical data mgt depts who filter through the data for missing and discrepancies.  These issues must be resolved before passing on th Biostatistics for evaluation, however, this data cleaning is ongoing as the data comes in-house on a daily schedule. During the clinical trial the pharma safety dept is required by law to report serious AEs to the FDA within hours of their knowledge of it.  I certainly want these guidelines to be enforced as a patient. 

High cost of drugs - - somehow they do it in other countries......  Drug development and research costs money to use the best educated and experienced medical researchers, teams, and latest most sophisticated laboratory equipment.  Better to have guidelines to prevent endangering patients lives IMO then to be slack and rush through a dangerous chemical compound. 

jenrio

cp418,

You are always a wealth of information/insight.  Thanks!

My main problem with delaying approval till OS comes out means:   Some patient need to make this decision to stay on the clinical trial, with reasons to suspect that they may be on the control arm, and possibly die when they might have lived with the highly promising drug (say TDM-1).    They have the choice of dropping out and getting on some compassionate access trials.    Then the OS data is not good anyway.

I don't mind pharma to make money with me-too product.   A lot of things need to be tried.   Now there are dozens of different conjugates in trial with different antibody+toxin combos.   That's good.   Even if TDM-1 is less effective than 99% on only Her2+ patients, but as a proof of concept, it deserves doubling down for investment and the inventor/investors deserve a healthy payoff.   Hopefully there will be more and more, for other antibodies with triple negative and for Her2-low folks.   Healthy competition and innovation lead to the cure for AIDS/diabetis/bacterial disease, I do not see cancer as being any different.

The best thing FDA can do is to allow alternative end-point and fund innovation that speed up the clinical trial process.  and there should be an xprize for startrek tricorder for CTC:

http://www.bbc.co.uk/news/technology-16518171 

 

cp418

 

FDA Considering Letting Some Breast Cancer Patients Try Experimental Drugs Earlier.

The National Journal (6/4, Subscription Publication) reports, "The Food and Drug Administration is considering letting some women with breast cancer try out experimental drugs earlier in the course of the disease, when they have a better chance of being helped, Reuters reports."

        Reuters (6/4, Steenhuysen) reports that, during an interview at the American Society of Clinical Oncology meeting, Dr. Richard Pazdur, director of the FDA's cancer drug office, said, "We're looking at introducing drugs into a very early stage of breast cancer, where a patient has a primary tumor and the chemotherapy is given before surgery." Dr. George Sledge, co-director of breast cancer program at the Indiana University Simon Cancer Center and immediate past president of ASCO, said, "We clearly need new approaches to getting stuff to the early-disease setting." The FDA's new approach "recognizes that we want to try to move drugs up front into an early disease setting as soon as possible -- not wait until the patient is beat up and out of chances and out of hope before we throw a drug at the patient and hope it does something."

http://www.reuters.com/article/2012/06/03/us-cancer-trials-fda-idUSBRE85208720120603

http://nationaljournal.com/healthcare/fda-considers-change-in-breast-cancer-drug-testing-20120603

jenrio

Interesting.  Quote from the reuter article:

"Researchers would then compare response rates in the two groups, and if the drug has helped cure significantly more cancers, it would be given a provisional type of approval called accelerated approval.

Drug companies would continue to follow women for several years to see if their cancers come back, and if women continue to be disease-free, companies could get full approval for the treatment. " 

So they are using neoadjuvant cure rate as basis for accelerated approval.   

http://ispy2.org/participants/trial-locations

Good FDA is listening and trying to be faster!   Currently the process is way too long and too expensive.    ISPY-2 is a good start, hopefully it would really speed things up.     But to go on warp speed, we need a good technology to monitor metastasis (or solve the 1-milliliter problem in radiology) and use it as alternative end point.

More information from ISPY-2:The I-SPY 2 TRIAL employs a groundbreaking clinical trial model that uses genetic or biological markers ("biomarkers") from individual patients' tumors to screen promising new treatments, identifying which treatments are most effective in specific types of patients. In addition, an innovative adaptive trial design will enable researchers to use early data from one set of patients to guide decisions about which treatments might be more useful for patients later in the trial, and eliminate ineffective treatments more quickly.

This large-scale trial involves a unique collaboration by scientists from the National Cancer Institute (NCI), FDA and nearly 20 major cancer research centers across the country. Study results will be made broadly available to the entire cancer research and development community.

...

The I-SPY 2 TRIAL will focus on treatment in the neoadjuvant therapy setting, in which chemotherapy is given to patients to reduce tumor size before surgery. All patients will receive the current standard of care and most participants will receive one investigational drug. A distinctive feature of the trial is that it will screen multiple drugs from multiple companies-up to 12 different cancer drugs over the course of the trial. In order to do this, FNIH received a master Investigational New Drug (IND) approval from the FDA-which allows the I-SPY 2 TRIAL team to graduate, drop and add drugs seamlessly throughout the course of the trial without having to stop the trial to write a whole new protocol. This will dramatically reduce the time it takes to move from one drug to another in the trial