NIH CLINICAL CENTER GRAND ROUNDS
Episode 2009-021
Time: 1:14:28
Recorded September 9, 2009

Contemporary Clinical Medicine: Great Teachers Treatment of Early Breast Cancer

Norman Wolmark, MD
Chairman, National Surgical Adjuvant Breast and Bowel Project Professor and Chairman, Department of Human Oncology, Drexel University College of Medicine
Chairman, Department of Human Oncology, Allegheny General Hospital

ANNOUNCER:  Discussing Outstanding Science of the Past, Present and Future - this is NIH Clinical Center Grand Rounds.
(Music establishes, goes under VO)

ANNOUNCER:  Greetings and welcome to NIH Clinical Center Grand Rounds, recorded September 9, 2009. This is a special Contemporary Clinical Medicine: Great Teachers lecture today, in which we'll learn the indentity of the winner of the 2009 Distinguished Clinical Teacher Award.  Our speaker today is Dr. Norman Wolmark, Chairman, National Surgical Adjuvant Breast and Bowel Project Professor and Chairman, Department of Human Oncology, Drexel University College of Medicine  Chairman, Department of Human Oncology, Allegheny General Hospital.  We take you now to the Lipsett Ampitheater at the NIH Clinical Center in Bethesda, Maryland, where Dr. John Gallin, Director of the Clinical Center, will introduce today's program.

GALLIN:  Welcome to Grand Rounds, and welcome to the first Great Teachers Grand Rounds of the year.

As tradition has it on this occasion, we also use this as an opportunity for the clinical fellows to make the announcement of their selectee for the NIH Distinguished Teachers Award.  This is an award that's become in my way of thinking one of the most highly sought after awards because it recognizes the finest qualities we hope to instill in our folks of mentoring, teaching and patient care.  And I invite Grace Chen who is head of the clinical fellows committee to make the announcement.  Dr. Chen.

CHEN:  Thank you.  The Distinguished Clinical Teacher Award was established in 1985 to acknowledge the exceptional work of our faculty and staff and it's one of the highest honors bestowed by the clinical fellows on an NIH senior clinician.  This board recognizes excellence in mentoring, healthcare professionals, excellence in issues relating to patient care and outstanding contributions to advancement of clinical research.
Today we are honored to recognize some of the top clinical researchers and mentors here at NIH.  Of course, we recognize and also thank all of our outstanding clinical staff for their intended commitment to training the next generation of clinical investigators.
A clinical mentor embodies the characteristics of an ethical and compassionate physician.
They anticipate road blocks and overcome op stack wills, all of the nominations this year that we receive from our clinical fellows should be recognized for their outstanding clinical mentorship, however, in particular, I'd like to thank this year's nominees, which are Dr. Jose Apus, Dr.Heller, Dr. Pamelor, and Dr. Maria Turner for their commitment to training clinical investigators at the NIH.  And I would like to thank the people who nominated these individuals.

Before we move onto comments about this year's winner please first join me for applause to thank and congratulate all of these outstanding clinical mentors.
[applause]

CHEN:  A few comments from our fellows about this year's award winner include that I can say without any reservation or equivocation that this is one of the finest teachers I have had the pleasure of learning from a mentor that clearly demonstrates a passion for teaching, a mentor whose clear respect and passion for patient care is only one of the many things I have had the pleasure of observing.  She exhibits this effortlessly one on one with the clinical fellows as well as to the clinical and medical students during rounds.

A teacher who responds to feedback about rotations and courses, implementing appreciable changes that have strengthened our training program in real time.

A physician whose intelligence, consideration for others and a sense of humor makes for an approach to leadership style that's gentle yet highly effective and as a consequence is clearly respected by faculty and trainees alike.  A physician who teaches by example, demonstrating compassion for patients and professionalism with peers, a role model who is confident for traits that make for a successful leader and finally this physician who is an indispensable attribute to NIH.  And makes it stronger through advocacy for fellows.

As a physician, this person teaches students and physicians at all levels. If these testaments were not enough, her background and credentials are incredibly impressive as well.

This year's awardee graduated cum laude from Harvard.  She received her medical degree from University of Virginia Medical School.  She subsequently completed an internal medicine residency.  She conducted bench research in virus diseases before returning to her clinical roots.  She has authored many textbook chapters across the spectrum of infectious diseases ranging from micro bacteria to small pox vaccine. 

She joined the NIH Staff in 2005.  She was a staff clinician in the laboratory of clinical diseases.  She subsequently came deputy hospital epidemiologist in 2007 where she directs the development, organization and implementation of the hospital infection control.  In this role, she has led several hospital-wide outbreak investigations, all the while continually educating staff with formal and informal instruction.

Better yet she sets an outstanding example.  To quote again from a statement of support, she excels in this position which broadens her role as a teacher beyond the fellow level to a clinical wide teacher.  And for this we are deeply privileged to have the honor and ability to opportunity to work with her.

Please join me in congratulating our 2009 award winner Dr.Tara Palmore.

[applause]

CHEN:  Tara is here.  Perhaps you could join us up here for a moment.

[applause]

CHEN:  Now I'd like to introduce Dr. Steven Rosenberg chief of the NCI Surgery branch who will introduce our great speaker this year, Dr. Wolmark.

ROSENBERG:  My congratulations to Dr. Pamelor as well, and we are about to hear from another great teacher, Dr. Norman Womark.

Dr. Wolmark is one of the giants of modern oncology.  I had the pleasure of working with him for a couple of years right after his training. Dr. Wolmark did his undergraduate and medical degrees at McGill university, and a residency in surgery at the University of Pittsburgh and it was shortly after that that Dr. Wolmark came to the NCI in 1976 to begin to do research in the surgery branch.

Norm and I spent two wonderful years together as we began our efforts to develop immunotherapies to the treatments with cancer and Dr. Wolmark was in on the first efforts to develop these therapies. Dr. Wolmark has gone on to a very distinguished career.

Shortly after leaving the NCI he became an executive medical director of the national surgical adjuvant project for breast and bowel cancers clearly one of the most influential groups that existed in this country. He has had a distinguished career rising to the position of director of surgical oncology, the surgeon and chief at the Montefiore University Hospital and in 1989 he became a professor in surgery and endowed chair at the University of Pittsburgh as well.

Dr. Wolmark has held many positions both in clinical care and in the administration of medical care, and in the development of clinical trials. He was the director of the Allegheny General Hospital and in 1994, he assumed the very important position as the chairman and principle investigator of the NSADP which he has held to the present time for the past 15 years.

He is in addition the professor and chairman of the department of human oncology at the Drexel University School of Medicine, a position he has held since 2001. The National Surgical Adjuvant Breast and Bowel Project is the finest example of the ability to bring physicians together to do clinical trials and the accomplishments of this are extending contributions to our understanding of the clinical care of patients with breast and bowel cancer.

I have the particular pressure to welcome Dr. Wolmark here as a great teacher and lecturer.

[applause]

WOLMARK:  Well, thank you for your kind invitation and thank you, Steve, for the introduction.

It is a home coming since, you know, my formative years in research took place, you know, on the tenth floor of this building, I think it was the tenth floor.  When I was originally contacted and asked to participate in this series, I realized that there were just some requests to which one could not say no. And this was certainly one of them.

The NCI has had such an enormous contribution and impact to the NSADP, that to have refused this invitation not only would have been crass, it would have been obscene as the slides suddenly disappear probably indicating that what I said was not appreciated by somebody.  Throughout, you know, the 50 years of the NSADP and we celebrated our 50th year anniversary some two years ago after we had randomized into our clinical trials 100,000 patients, throughout those 50 years we have been funded by the national cancer institute originally by the national institute of health when it was a singular institute and then subsequently throughout our existence we have been funded by the national cancer institute. And without the NCI, we would not have existed. That is, it's that simple.

So in essence, the relationship that we have had with the NCI has been a very close one.
It has been like a marriage.  It has had its great times with endorphin rushes, it has had its good times and, of course, it's had its other times.  And yet, we are linked so I am proud to be here not as a great teacher, which I don't think I am, but to pay homage to the NCI
And for having the courage to invest in the clinical trials process as embodied by the work of the NSADP

So on the first slide, in compliance with CME, we disclose our transgressions.  And based on your definition, I am pleased to declare that I have had no conflicts, but those of you who know me will also conclude, and understand that at one time or another, I have been in conflict with just about everybody.  So, of course, it depends how you define the task, which is approve clinical trials as well.  They are also obligated by CME. mandates to set out objectives because without setting out the objectives, you probably would wouldn't understand what this lecture is all about.

So let's go ahead and formulate some objectives, which perhaps you will find non-traditional.  One, to determine the role of clinical trials in changing the standard of care of early stage breast cancer.  Was this effort worthwhile?  Two, to illustrate the pivotal function of the NCI in advancing the state of the art through clinical trials.   Was the investment worthwhile?  Was the investment in the NSADP and other cooperative groups worthwhile in changing the standard of care?  And then in this era, in this era of posterity, in this era of a decreased budget and funding, we have to examine the scenario that none of us wish to examine.  Since this is a marriage, what would be the consequence of a potential failed marriage.  How far are we removed from the days of tyranny, the randomized prospective clinical trials represented the clinician's laboratory, the opportunity to advance the state of the art in a disinterested and objective manner.

It was a mechanism whereby we could challenge what our forefathers taught us, the randomized prospective clinical trial was to end the era of tyranny, the era of authoritarianism, what an individual armed with a retrospective case series and based on his personal charisma could ascend to the pulpit as a great teacher or otherwise, and with silver tongue powers could dictate the way a disease is treated for generations and generations to come.  How far are we from that age of tyranny?  Many of us complacently think we have left it behind.  In perpetuity, we have not.

The ecology is a very fragile one, and we are perilously close to returning to that era of tyranny.  So moving right along, the NSADP Clinical trials have been instrumental in the retreat from radical mastectomy in establishing the breast preserving operation as a standard of care for primarily operable breast cancer.

Now, a surgeon who looks at this certainly has to come face to face with one of his worst nightmares.  Because, of course, taken to its extreme and logical conclusion, the surgeon has to come to grips with the fact that perhaps some of us are now regarding the operation, challenging the primacy of the operation and perhaps regarding the operation as the adjuvant rather than the dominant intervention, an adjuvant to another intervention such as chemotherapy and herceptin.  That will become the primary intervention with the operation as the adjuvant.

Do we have enough date to support this radical hypothesis and thinking?  And yet the surgeons of the NSADP have embraced this concept and evolved a series of preoperative adjuvant trials.

Let's not bandy about, let's cut to the chase and let's formulate the pivotal question for this discussion, which I did not show on these specific aims of the obligatory slide.
Have we crossed that elusive threshold between therapy based on empiricism to therapy based on well defined molecular markers?
 
Have we crossed the threshold of therapy by phenomenalogy to therapy based on enlightenment, again, using well defined molecular discriminates and as target and a method to define populations who will and will not benefit from a particular therapeutic intervention.

So with that as the guiding question, let's proceed.

Whenever one has the temerity to challenge the value of the radical mastectomy, one must play homage to William Stuart Halstead shown in an unusual pose in a photograph given to me by one of my mentors, mark ravage who's only a few professorships removed from William Stuart Halsted showing some humanitarian properties not commonly associated with the man such as his love of dogs and puppies.
Perhaps we can pick up on this canine scene later on in the presentation if sufficient time remains.

It is important to ask how Halsted views the dissemination of solid tumors and in particular breast cancer, because if we understand that, then we will have an appreciation for why Halsted evolved the radical mastectomy and more importantly the whole concept of resection.  Halsted believed that breast cancer and solid tumors disseminated through local regional pathways from the primary tumor cells would percolate along lymphatic to involve the regional node.  From the regional node, cells would travel along the fascia that envelopes muscles to involve distant organs.  That metastasis arrived at the liver.

The surgeon's infatuation with the fascia dates back to this particular hypothesis, what is conspicuously absent from this understanding is any role for hematogenous threads.
He knew it exited but stated that it occurred rarely, if ever, in the dissemination of breast cancer.  So clearly if this is your biological perception of the disease, it was entirely logical.  In fact, obligatory for Halsted to involve an operation that embraced the principles, namely the radical mastectomy and on block resection, which is comprised of three components.  The extirpation of the organ.  And the removal of all intervening tissue within a fascial envelope including muscles because the fascia was a conduit for the dissemination of tumor cells.

Now, this hypothesis was limited to breast cancer.  It was not limited to the radical mastectomy.  In 1908 when earnest miles resurrected the one step abdominal peroneal resection for one reason and one reason only, quote, to keep up with the splendid gains that have been made in breast cancer by William Stuart Halsted.  It is the documented biologic rational for radical head and neck resection.  It is the documented biologic rational for radical resection of melanoma.  It has the biological rational for the radical hysterectomy and it would have been the documented biologic rational for radical prostatectomy were urology endowed with the literature. 

How was it then that Halsted was able to convince colleagues, his co-evils of the value of the radical mastectomy.

This is a photograph that was taken from the 1895 issue of the annals of surgery.  He used the techniques that were available to him at that time.  The anecdotal respective series, and here Halsted lists the local regional recurrences of the then principal surgeons of the day.  82% local recurrence.  Churning, 62%, in an uncharacteristic display of humility, Halsted lists himself last with a 6% incidence of local regional recurrence which must have been absolutely astounding to the then principal surgeons of the day.

So we also must ask were there no more outcry?  Was there no resistance?  Did somebody stand up and say I disagree with you?  In point of fact, somebody did.

In 1998 Halsted addressed the American Surgical Association in New Orleans.  And at the end of the discussion, he stood up and thanks the audience for uniformly embracing his biologic hypothesis and his operation.  We could only derive two conclusions from Halsted's comments.  He was either asleep or he was under the roof because during the discussion that followed Halsted's presentation, some of the most far reaching  observations were made relative to the biological behavior of breast cancer, and they were made by this individual Rudolph Matti of New Orleans, better known for the treatment of abdominal aortic aneurysm.  Who in this room has heard of Rudolph Mattis.
Ladies and gentlemen, he has been purged from the historical annals of breast cancer.
We discovered his quote by accident. 

When Mark Ravage was reviewing the minutes of the American Surgical Association for his book A Century of Surgery, they were suppressed.  Rudolph Mattis stood up and criticized Halsted and said, quote "a far more important question of that of local recurrence isn't that which relates to the actual number of cures obtained by present methods.  You are using the wrong end point."

He further went on to admonish Halsted, hence the words constantly used to qualify the operation of the day is complete and radical from the latin radic, anatomic misnomers which serve solely to indicate surgical technique and are elusory if used in the sense that they root out the evil with any degree of certainty.  Mattis understood that breast cancer was a systemic disease.  That patients were dying not because attention to inoperative detail but because the tumor was disseminated hematogenously, but, alas, the voice from Hopkins was far from persuasive, far more eloquent, far more authoritative than that solitary whisper from Tulane, and Rudolph Mattis has been eliminated from this story and almost forgotten.

It was not until 70 years later, and here, 110 years ago, we have framed the challenge of crossing the threshold, the scalpel versus the hypodermic needle.  And it's not until 70 years later that Bernie Fisher resurrected these two mutually exclusive hypothesis, but resurrected them not with the intent of debating them ad nauseum, but of resolving these two hypothesis using the scientific method, the randomized prospective clinical trial.
I took this photograph of Bernie Fisher in 1973 when I first joined, and he looked up at me and he said, Norman, don't take my picture.  I said, well, why not?  He said because some day you are going to use it to embarrass me.  Well, perhaps some day I will.
Today is not the day.

So though Bernie Fisher made the commitment to use the scientific method the randomized prospective trials to address these two mutually exclusive hypothesis which were formulated at the end of the 1800s.  And his effort was initiated in 1971 with NSADP  Protocol, a protocol that had an enormous impact on the surgeons of today, but today is all but been forgotten.  It is not indexed.  You can't look it up electronically, we could argue it didn't exist.

So what was magical about life in '71?  The effort started in 1971.  Was that an important year?  Well, in point of fact, it was an important year because it ushered in an age for the NCI  and ultimately for randomized prospective clinical trials because with typical American bravado, Richard Milhous Nixon declared war on cancer signing the bill that would provide funds for the NCI and to support the clinical trials mechanism.  This was a pivotal and important event.  Bernie Fisher still has a pen from the signing of the bill.

Why did he keep the pen because many pens were used to sign the bill.  I think we got the Bic pen, it wrote upside down but nonetheless it was an important symbolic reminder.
The NCI  had intrepid, courageous leadership under the direction of Vince DeVito who was always able to obtain sufficient funds from congress to support clinical trials and put an enormous premium on them.  In 1974 he moved the cooperative groups into the division of cancer therapy, and the age was further stimulated.  So this was a halcyon time for the randomized prospective clinical trials.

So let's get back to bo4.  They clinically took patients and they were randomized through the Halsted radical mastectomy or a total mastectomy or a total mastectomy and radio therapy.  Close to 1100 patients were randomized into these studies.  What is a total mastectomy, you ask?  Well, a total mastectomy is really a simple mastectomy.  As surgeons we don't like to use the word simple because we don't know whether it refers to the operative procedure or to the individual performing the operative procedure.  But a total mastectomy leaves the muscles, the fascia, and the lymph nodes behind completely unvulnerated.

This certainly would have perturbed Halsted in a significant manner.  Although we thought that these patients were clinically negative, as it turned out 40% of the radical mastectomy group had histologically positive nodes.  Because this is a randomized trial, we argue that the total mastectomy group also had 40% histologically positive nodes although they were left behind completely untreated so here we have an opportunity to test the two mutually exclusive hypothesis, and we published the 25?year update from this trial in the New England Journal of Medicine in 2002, the original report was in 1977 showing that the three lines are superimposable.  This had an enormous impact on the surgeons at the time.  It convinced surgeons that altering operative nuance is not likely to result in significantly increasing the cure rate.  And if one wished to do that, one had to use systemic intervention for a systemically disseminated disease.

So the retreat from radical mastectomy and the ascent of adjuvant therapy including biologic are inextricably intertwined.  It also provided the biologic basis for the next step in the retreat from radical mastectomy, and that the rationale for the breast preserving trial NSADP protocol bo6 into so bo6 started in 1976 and was simply an extension of bo4 where patients were randomized.

Try and imagine putting this in front of an i.r.b. today.  Patients were randomized to the radical or modified radical operation of breast preserving procedure with an axle layer y node dissection and no further treatment. The breast preserving procedure axle layer y node dissection 20 which therapy to the lateral breast was delivered and 8,851 patients were randomized in to trial.

So what was the value of radiation?  What was the incidence of breast recurrence following lumpectomy?

So in 1984 we submitted a manuscript to a very prestigious journal that shall go unnamed which is located on number 10 Shaddic Street, Boston, Massachusetts.  And we waited for a solid year for the critique to come back.  A year. Had I not witnessed this myself, I would not believe it.  We called the editor every second week and were told gentlemen, the critique is in the mail.  We waited another two weeks, no critique.  Vince DeVito called and was given the same story.

Finally the blessed day arrived.  A manila envelope is present, and we all gathered around the table to determine what the critique was, what is it that took us all of these years to formulate, and with trembling hands we tore open the envelope and read the critique.
And to our understanding it appears that they and they said, quote, gentlemen, breasts don't recur. Tumors do.

Well, all right, granted, but was this the effort of one year's deliberations?  Clearly the surgeons who were on the review panel found this far, far too provocative to accept as a manuscript and it was delayed for a solid year. Ironically, if one read that's manuscript very carefully one will see the term breast radiation used throughout.  Now, we know that breasts don't radiate.  Well, perhaps some do, but not in this traditional sense.

So what is the role of breast radiation?  So we had to change the term to ipsilateral breast or i.b.t.r.  To be sinatically correct.  While we were waiting for this manuscript to be reviewed an interesting phenomenon occurred.  We didn't understand the significance.
The great state of California passed legislation through the state house that mandated that every patient with breast cancer must be apprised of alternatives to radical mastectomy and in particular to the value of radiotherapy unless the governor reviewed our manuscript, he did this completely unencumbered by data.

So here we have an example of political expediency coming into conflict with data that was generated using the scientific method and we were naive.  We had no idea what the significance of this was.  We simply passed it off as being California.  And we know that politics in California are incomprehensible and inexplicable.  Then, now, and forever more.  We paid an enormous price for our naivety, as we shall see.

So ipsilateral breast recurrence, what was the value of the radiotherapy, and we hypothesized early on that the group that did not receive radiotherapy would have a much higher incidents of ipsilateral breast tumor recurrence and in fact they did.  We published the 20 year update in the the New England Journal again in 2002.

We see that 40% of the non-irradiated lumpectomy patients had an i.b.t.r.  Compared to 14%, but we had hypothesized a (...) based on the results of do4 that survival would be identical in all three groups and when we published the survival data initially in 1985 and in the 20-year update, we see that comparing lumpectomy without radiation to mastectomy, no difference.  Comparing mastectomy to lumpectomy plus radiotherapy, no difference.  And the alternate hypothesis once again supported, and the flood gates to adjuvant therapy were opened.  And it finally did come.

If you participate in provocative clinical trials which challenge the standard of care and the very basis that is used to disseminate therapy, you are highly visible.  And in 1994 we were investigated by a congressional subcommittee because an individual in montreal submitted fraudulent data and it nearly destroyed the NSADP.  A good day for us was the day without a subpoena.  I don't know how to describe to you what it feels like to be on the receiving end of such an investigation.  But perhaps I can give you an analogy, which may or may not be appropriate.  It's like being examined for a polyp that's 30 centimeters with a 25 centimeter sponge.  No doubt there will be those who will enjoy the journey, but it is unlikely to result in unsubstantive accomplishment.

So the flood gates to adjuvant therapy were thrown open in quick succession with the support of the NCI, we demonstrated as early as 1975 that giving patients with positive nodes two years of l path, two years of oral l path, we were the surgeons, we got the oral preparation because they didn't think we could use iv chemotherapy and showed proof of principle that it had the potential to save lives with a statistically disease free survival advantage.  This, of course, was quickly eclipsed a year later by the CMA data.  And it's the credit to the NCI that supported that.  Showing a statistically significant advantage in favor of CMF over no CMF.

Again, proof of principle.  And, then we just had a process that continues, and continued as part of the age.  The 70s brought us cmf and tamoxifen through do9.  This subsequently led to d14, tamoxifen in tumor patients.   It led to randomizing 30,000 patients to two prevention trials which established tamoxifen as an approved chemotherapy agent.  The 80's brought us the recycling of the psych of the of the NSADP now thought to be anachronistic.  It brought us chemotherapy trials where chemotherapy was given before the operation for stage 1 and 2 to see if we could minimize the scope of operative intervention.  The 90s brought us the taxing albeit quick succession and I have no intention of giving you encyclopedic anthology to render the audience comatose.
The variations in sequence of chemotherapy, taxing, either giving them sequentially or together in the form of tax, it brought us the era of chemotherapy and today we are debating whether we took a wrong turn in the 90s, and whether we need anthracycline  particularly in 02 negative breast cancer and we are comparing together with u.s.
Oncology, tax, adriamycin.

There is no substitution for the trial, although there are enormous number of papers that have been published speculating that an atlanta sigh cleans are not necessary in her 2 negative breast cancer because herceptin is used in that setting all speculation.  So the preoperative trials which were very courageous because here we had the surgeons pursuing a therapy that would limit the impact of their operative procedures.

D18 compared four cycles of ac pre-op to four cycles of ac post-op.  To determine if pathologic complete response was empiritive of the chemotherapeutic regimen.

D27 added ac and doubled the pathologic complete response rate from 13 to 267%.
Our two current trials which is accruing reasonably well are very provocative because we are using pathologic complete response as the primary end points as a surrogate of the efficacy of a regimen.  In d40 we are using doubling of chemotherapy plus or minus, and if the chemotherapeutic regimen has not achieved, a significant pathologic response advantage they will be abandoned.  In essence it is an invivo bioassay.

For her 2 positive patients we are comparing the combination of herceptin.  Here we have the opportunity to determine the preservation of the tumor as it is being watched, as it is being sampled.  It is a way to develop a molecular profile to determine the value of a particular intervention and we see that those patients who achieve pathologic complete response in those protocols of the nsadp is done using preoperative therapy do far better than those who do not achieve a pathologic complete response.

So clearly, this is an enormous opportunity to develop amolecular taxonomy that will determine the value of a particular intervention.  We have made enormous efforts to achieve that.  We used a 50 gene risk predictive index on an assay gene chip using paraffin embedded tissue and it was a difficult process.  We were far more successful in determining the outcome of those patients who did not achieve the pathologic complete response compared to those who did.  For those who did not achieve the pathologic complete response, based on this 50g profile, we were able to divide them into a low risk and high risk group, we were not as successful in determining who would or would not develop pathologic complete response.
Certainly an endeavor that is very, very much needed in this day and age and which we are diligently working on. 

So up to this point, I have assiduously avoided addressing the pivotal question that I started have we crossed the elusive threshold to therapy based on molecular.  And is this a real phenomenon and many of us think that may of 1985 was that pivotal example when we did cross the threshold, when the herceptin adjuvant data were presented or is it simply a naive misguided idealistic hope that had absolutely no basis in reality?  Are we simply deluding ourselves?   If we look at some of the authoritative scientific publications available in this country such as US News and World Report, one would think there is no debate that we have in point of fact crossed that threshold, and crossed it definitively, and we crossed it on May 16th, 2005 when the herceptin data were presented and within this issue they try and describe the poignancy, the emotion, the excitement that existed in a room crowded with 12,000 people in Orlando when the NSADP and north central data were presented.  And I was in the room and turned around and noticed that a significant proportion of the audience had tears in their eyes.  The crowd applauded loudly.  The overflow audience in the Orlando hall was clearly grateful it was a rolling stone in terms of response.  Truth to tell, I had a hard time - a well-respected oncologist from Indianapolis.  We know that midwesterners don't change either at the height of ?? and yet we have George Sway describing these things with, well, whenever you get this you are always well advised to get another opinion.  And we did.

We submitted the data from the combined analysis that the north central to the New England Journal of Medicine.  And we waited for the tyranny of the editorial because no matter how many years we spend on carrying out clinical trials, carrying it out with two paragraphs of an editorial, you can have your work completely (...) so we waited.

The editorialist who gave Anderson, an individual who is not prone for his hyperbole, a metaphor, what is it that he said about these data?  The results are simply stunning.  Beautiful women are stunning.  I mean, only a buy we statistician doesn't make contact in a log rank analysis and call it stunning.  And here he is using stunning.  Clearly the results were reported in the issue are not evolutionary, but revolutionary.  Further, the two reports in this issue were completely altered our approach to the treatment of breast cancer.

We didn't pay him.  Otherwise I would have disclosed it right at the beginning.  So here clearly both US News and World Report and the New England Journal believe we have crossed the that elusive threshold so, in other words, if we were to believe these two publications in a little over a hundred years we have gone from this to this that William Stuart Halsted rescued to an epiphany, the rolling stones concert so when all else fails think one is obligated to examine the data in a disinterested and objective way in 1987 ? he also concluded that these 79 women had a worse prognosis than women who did not amplify her 2 new.  This led to the development of a humanized monoclonal ant body named trasdusamed or herceptin.

And it was shown that it was effective in advanced disease in patients who were hur2 new positive and starting in February of 2000 there were four well conducted clinical trials with 14,000 patients assessing the value of herceptin in the adjuvant setting.  And the north central 9831.

Since these trials were very similar, we combined the common arm, and had 3351 patients that were analyzed for the value of adding herceptin a.c.  Followed by taxal, a.c.
Followed by taxal to which herceptin was concomitantly and continued for a total period of one year.  And in 2005, at a prescheduled meeting of our data monitoring committee, at the first interim analysis, I was shown this swatch.  Perhaps he was right.  Perhaps, the data are stunning because the differences is attributable to herceptin were enormous.

You can wait a lifetime to see data with this kind of magnitude of benefits.  In favor of herceptin at four years of absolute difference of 18% the event rate was halved.  A hazard ratio of 0.5.  It had 11 zeros after the decimal place, 11 zeros after the decimal place and we are argued with the editorial board of the New England Journal.  They told us we could only let you put three zeros in.  We said why, there are eleven zeros, they said that's our policy.  We say well how many trials have you seen with eleven zeros after the decimal place, they said they had none.  There were three zeros.  They wouldn't budge. So there is a striking benefit associated.

You say, well, it's early but these are the definitive results because at this point when the data were disclosed we made herceptin available to the control arm so these differences really reflect the best that we can determine associated with herceptin.  There was also a survival benefit, although this was not the primary end point.  So, this was the data that we showed in May of 2005.  They have been updated, but these control patients did have herceptin, significant proportion of them.  Edith Perez updated these data, and you still see a hazard ration of .5 with a highly significant p-value, not 11 zeros, but it's less than, and a statistically significant survival advantage.

Certainly this is a clear example that we have crossed the threshold between therapy by empiricism, the therapy based on molecular discriminants.  And, based on these data, the FDA approved herceptin for the adjuvant treatment of her-2-positive, node-positive breast cancer.

Do we really understand the mechanism?  We can easily become complacent.  We look at data, we say yeah, you know, her-2 is being targeted; that's the end of the story.  But when we presented this data we also examined the data according to her-2 status, although we specified that all patients had to be her-2 positive.  When soon pake [spelled phonetically], who was head of our pathology division at the NASDP, reviewed the result centrally, we learned that there was a proportion of patients who were, by classic standards, her-2 negative.  In particular, there were 174 who were fish-negative and ihc less than 3, which by all definitions, her-2 new negative.  And yet, if we look at this plot, the forest plot, we see that they benefited from herceptin to the same extent as her-2 positive.

So perhaps the definition that applies appropriately to advanced disease may not be the definition that determines efficacy in the adjuvant setting.  And these differences for the her-2-negative population are not trivial.  If we look at the kaplan-meier plots for this her-2-negative population that was evaluated centrally, we see a relative risk of.34.
This is, you know, a 66% diminution in event rate for a her-2-negative population attributed to herceptin with a statistically significant p-value.

So we, of course, quickly submitted a concept to the NCI, who supported it, to their great credit, for assessing herceptin in the her-2-negative setting.  Of course, we needed Genentech to be part of this, and we agreed that we would go through a process of having these slides reviewed, actually, having the blocks recut, submitted to three independent pathologists to determine if what we centrally were calling her-2 negative was in fact her-2 negative.  This study went through two IRBs, and the link between the results from the three independent pathologists and the data resided with one individual, an honest broker who could link the data with outcome.

Fortunately, we were able to recruit this honest broker from the comptroller's office at Enron.  And by way of an update, I can disclose to you that he's currently retired in an orange jumpsuit inspired by Carl Lagerfeld with accessories by Marc Jacobs.  So, what did these three pathologists come up with reviewing these so-called negative data?
Well, a pretty well identical result.  These curves are almost superimposable: relative risk of .3, a statistically significant p-value.  What was the level of concordance, overall, between the centrally reviewed and the three independent pathologists?  That the level of agreement was over 95%.

So, clearly this trial, which certainly challenges the envelope, in a her-2-negative population that has been approved by CTEP, has taken two or three years to put into place where it is in a position to get started, I think, will determine definitively whether herceptin in the adjuvant setting is or is not effective for a her-2-negative population.
Tc, taxotere cyclophosphamide, tc, taxotere cyclophosphamide plus herceptin.  So, clearly, when we think we know the whole story, we're well advised to reassess it in a definitive and objective way.

Well, we've also targeted patient populations to determine, based on molecular profiling, who will and will not benefit from a particular intervention.  The NSADP, together with genomic health, developed the oncotype dx assay for node-negative, receptor-positive breast cancer.  The reason we were able to do it is that, from trials that were 15 years old, we had an annotated tissue library to be able to validate a punitive algorithm, and without validation they are a dime a dozen.  You will see them in every publication on a weekly basis.

So, the algorithm consisted of 16 genes that were associated with outcome in univariate analysis, five genes to normalize data because we were using rt-pcr from paraffin-embedded tissue that had been in storage for over a decade, and we validated this with nsabp protocol b-14 data, where we had 668 from tamoxifan-treated patients showing that 50% of the population had a good risk, 93% distant disease-free survival, compared to a high-risk population that had a 69% distant disease-free survival at 10 years. 

These are arbitrary cutoffs because, in essence, this is a continuous variable, and that is how recurrent score is used commercially.  A 20% recurrent score will translate to between a 10% and 15% distant risk of recurrence at 10 years.

So, clearly we validated this algorithm in a definitive way because we had an annotated tissue library.  That was just an hne [spelled phonetically] cross section of my amputated right middle finger.

So, here we ask whether this assay is applicable to the benefit of chemotherapy, and we were able to do this because we had annotated tissue from b-20, where we compared tamoxifan to tamoxifan plus chemotherapy, showing that 50% of the population that fell into the low risk showed no benefit from cmf-based chemotherapy.  The intermediate risk didn't show a striking benefit, but again, we had too few patients and that it appeared that the entire benefit from adjuvant chemotherapy that was cmf-based fell into the high-risk category, which accounted for a third of the patient population or a third or less.

So it appeared that the benefit was limited to the high-risk group.  This is probably true for node-positive patients as well, with the data from swog 8814 having been reported.
But again, the confidence in this analysis doesn't nearly have the power that it does for node-negative, and this led to the enormous ongoing trial supported by the NCI, the tailor x, where those individuals who were at low risk received hormonal therapy.  Those at high risk all get chemotherapy.  The intermediate-risk group is randomized to receive chemotherapy versus no chemotherapy.

Again, this is a tour de force, and a great deal of credit should go to the NCI for initiating this study, which is accruing extremely well.  7,650 patients out of a required 11,000 have already been randomized since 4 of '06 to determine whether the intermediate-risk group will or will not benefit from adjuvant chemotherapy.  Commercially, 90,000 assays have been done.  It's estimated that one-half of the population that's node-negative and [unintelligible] tumors are receptor-positive has this assay performed to determine whether they ought or ought not receive chemotherapy, with the intermediate group being an excellent candidate to enter into this protocol.

So, have we crossed that threshold?  And I think that we certainly have.  But I think we have to be vigilant, because I think the distance between the days of tyranny and enlightenment is much smaller than we think.  Without the continued support from the NCI, I think that there is an enormous danger to devolve into the tyrannical era of authoritarianism, where therapy could be dictated by individuals, which we think would be a catastrophe.

So, I started this lecture by disclosing to you some surgeons' nightmares.  Let me disclose to you one of my recurring nightmares.  And you may suggest that I receive psychotherapy, medication.  I understand that.

How will a guest lecturer from this lectern look back upon this era 15 years from now?
Will they look back upon this era and say it was populated by individuals who had the courage of their convictions, who had the intrepid fortitude to move on and to do what was necessary to move the state of the art forward?  Or will that individual be inclined to show this slide?  And I really worry about this, because I think this is not simply a hypothesis or an apparition.  I think this is a real possibility if support for clinical trials disappears and the marriage between the NSADP and the cooperative groups is dissolved or annulled.  Clearly we have the mechanism to make enormous progress.

It's up to us to apply it in a definitive and unequivocal manner.
Thank you very much.  I am thrilled and honored by the invitation, and I hope, Joanne, that our marriage will continue.

[laughter]

Thank you.

[applause]

ANNOUNCER:  You've been listening to NIH Clinical Center Grand Rounds, recorded September 9, 2009. You can see a closed-captioned videocast of this lecture by logging onto http://videocast.nih.gov -- click the "Past Events" link -- or by clicking the "View Videocast" link on the podcast homepage at www.cc.nih.gov/podcast. The NIH CLINICAL CENTER GRAND ROUNDS podcast is a presentation of the NIH Clinical Center, Office of Communications, Patient Recruitment and Public Liaison. For more information about clinical research going on every day at the NIH Clinical Center, log on to http://clinicalcenter.nih.gov. From America's Clinical Research Hospital, this has been NIH CLINICAL CENTER GRAND ROUNDS. In Bethesda, Maryland, I'm Bill Schmalfeldt at the National Institutes of Health, an agency of the United States Department of Health and Human Services.