Musings of a Cancer Doctor
Wide-ranging views and perspective from George W. Sledge, Jr., MD
Sunday, September 22, 2013
There’s been a great deal of talk lately about the price of cancer drugs: Op-Eds and articles in the New York Times, a slew of interviews of prominent oncologists in every imaginable media outlet, and impassioned editorials in leading medical journals.
The sense many of us (myself included) have is that the current system of drug pricing is ultimately unsustainable. How many $10,000-a-month drugs can a health care system sustain before collapsing? We’ll discover the answer in the near future. The prices are already well above what resides in the average uninsured American’s bank account.
Outside health care, the introduction of a me-too product (a new smartphone or computer tablet) results in a significant reduction in price, a race to the bottom. In PharmaLand, though, a new imatinib clone for CML results in an increase in price that far outstrips added benefit, accompanied by price increases for older drugs. Drug costs seem effectively outside the modern capitalist economy.
The economics of drug development and pricing are complex, and I do not claim to understand it, though I have tried. Bottom line: prices for on-patent drugs always increase.
One of my favorite Tolstoy stories is the wonderful “How Much Land Does a Man Need?” Its protagonist, the landowner Pahom, is never satisfied with the amount of land he owns. He moves from place to place in his pursuit of more farmland, finally leaving the settled areas for the wild places at the edge of the Tsarist Empire. There he meets a tribal chieftain who tells him that, for a thousand rubles, he can have as much land as he can circumnavigate on foot in a day.
After walking all day he sees the sun setting, and races back to where he began his walk, only to collapse and die at the finish line. The story ends: “His servant picked up the spade and dug a grave long enough for Pahom to lie in, and buried him in it. Six feet from his head to his heels was all he needed.”
Indeed. Is the pharmaceutical industry racing, Pahom-like, to an early grave, in its pursuit of wealth and new compounds? And how much “land” does it really need to innovate, to create the next generation of drugs, while satisfying its stockholders? Again, I don’t claim to know the answers. I only know that this can’t go on forever.
In Tolstoy’s Russia a man’s worth was based on the amount of land he owned. Today wealth has a different measure, something incomprehensible to our forbearers: insubstantial collections of photons hiding out in distant server farms, mirages we have collectively agreed to call “money.”
And that wealth, in turn, is capable of buying life, at least in small increments. Debates over drug costs are interesting because they revolve around value judgments. Consider this thought experiment: imagine someone hands you a glass of water and says, “This will cost you a dollar, has no side effects, will palliate your disease-related symptoms, and will prolong your life by 12 days” (the EGFR/pancreatic cancer scenario). Pretty much everyone would drink deeply. We value life when the price is low.
But the other knock on our new drugs is that while we pay an immense amount, we often get little in return other than side effects. EGFR inhibition for pancreatic cancer, with its 12 added days of life, is free neither of cost nor of toxicity. The old Woody Allen joke applies: two old ladies are talking about a restaurant they’ve just eaten at. The first woman says, “The food, the food was so awful.” Yes,” says the second woman, “and the portions, the portions were so small.”
And the further one gets away from the magical “glass of water” scenario, the fewer willing to drink the Kool-Aid. Where one draws the line is always a value judgment. Always. And it leads to the bitterest of arguments: remember “Death Panels” in the run-up to the Affordable Care Act? We are, as a society, deeply uncomfortable with telling taxpayers that their lives lack infinite value.
Cost, life added, toxicity. The health outcomes folks have wrestled with these three for years, with their Cost per Quality-Adjusted Life Years (or QUALYs, as they are know in the biz). QUALYs are an attempt to provide some quantification to an inherently unquantifiable ethical quandary. $50,000 per QUALY? $75,000? $100,000? $250,000? Are our QUALYs inflation-adjusted?
The Cost per QUALY equation basically asks us “How much is a life worth?” Patients, of course, are also interested in the related question, one little studied by health economists: “How much is my life worth living?”
My patients have given me many different answers to that question. Some are serene about the length (and end) of life, others terrified by onrushing oblivion, and some concerned about the fallout their passing will create. Sometimes their feelings overlap with religious beliefs, sometimes with age, sometimes with perceived responsibility. I’ve been a doctor long enough not to generalize about how patients answer the question, or even about whether there is a right answer. Three brief vignettes will suffice:
1. An elderly breast cancer patient had indolent metastatic breast cancer, which routinely shrunk with endocrine therapy whenever I could convince her to take it. But she was sad -- sad because most everyone she had cared about was dead, and her children were neglectful and far away, so she took the pills only irregularly. One day, as she left the clinic, I said “See you in three months.” With a withering look directed at me, she said, “God, I hope not.”
2. A woman in her 40s found that her husband was sexually abusing her son. She couldn’t convince the authorities to prosecute; not enough evidence to convict. She didn’t want to expose her child to the vagaries of the legal system, so she cut a deal with the monster, divorced, and got sole custody. She came down with a nasty, ugly, estrogen receptor negative breast cancer. It was the late 1980s, and I didn’t have much to offer her, but she would do anything to add even a day of life, so I beat her up with toxic and ineffective regimens. She looked exhausted the last time I saw her, but was horrified when I told her I had nothing else to offer. She wasn’t afraid of dying. She was terrified that at her passing her child would fall, once again, into the evil hands of her ex. I don’t have a clue what happened to that child: I still think about him from time to time.
3. An old West Texas small-town city councilman developed non-small cell lung cancer and came to the VA for care. I was a brand-new oncology fellow, and I offered him on an investigational chemotherapy trial. His tumor responded, quite nicely, for six months or so, but eventually progressed. Back then, like many novice physicians, I felt heroic when my drugs worked, and a failure when they didn’t. I apologized profusely for the progressive disease we saw on his chest x-ray.
He waved my explanations off: he was one of those stoics the high plains produce in quantity, and uninterested in my apologies. In fact, he ended up consoling me. “Let me tell you something, Dr. Sledge, about those six months you gave me. In my town there’s a poor section [he didn’t say it, but it was understood that he was talking about the Mexican-American barrio] that didn’t have running water. No one gave a damn. I always felt bad about that, but I went along with the rest for years. I knew I didn’t have much time left, so I rammed a water bill through the city council. Those folks will have running water after I’m gone.” He didn’t smile at all when he said this, but had a look of grim satisfaction on his face, the sort one see when someone has righted a deep and longstanding wrong.
We must base public policy on health economics, not on anecdotes, of course. The drugs are obscenely expensive, they don’t do enough, and they often harm quality of life at the end of life. They are used long past any reasonable chance of benefit. Resources are limited, and the decision to use expensive and often ineffectual drugs draws money away from other valid social and personal goals (spend all your money on co-pays for the latest RTKi, and forget sending Johnny to college next year: you’re broke). Drug pricing also distorts oncology practice, a practice bathed in perverse incentives.
Yet “how much is a life worth?” the Cost/QUALY equation, still leaves me a bit queasy. Queasy, because I don’t have a good answer to the question “how much is a life worth?” and because my patients have so many, and equally valid, answers to the question “how much is my life worth living?”
Will a person’s worth ultimately be based on the amount of life he can afford? A fascinating new book (The Book of Immortality, by Adam Gollner) describes how a group of billionaires are funding anti-aging research. We oncologists squabble over the meaning and cost of a few months of life, often earned with appalling toxicity. But these guys don’t care about cost (the server farms are full of their ghostly wealth-photons), nor do they care about a few crappy months: they are shooting for immortality, or something like it. The Big Time, literally.
The book mentions Larry Ellison of Oracle, one of our richest software moguls, whose biographer says that he sees death as “just another kind of corporate opponent he can outfox.” Not for Ellison the Bill Gates approach to philanthropy, buying mosquito netting and creating vaccines to prevent the deaths of real people in Sub-Saharan Africa. His Ellison Medical Foundation gives out more than $40 million per year for research dedicated to ending mortality, by “understanding lifespan development processes and age-related diseases and disabilities.” OK?
If a drug tripled or quadrupled the normal human life span (forget immortality: there are just too many asteroids, lightning strikes, paranoids with assault rifles and drunk divers out there) what would you pay for it? A lot, I suspect. Who would be allowed to use it? My bet would be on the billionaires, and the devil take the hindmost.
Science fiction is chock full of stories exploring this theme, but popular culture has a word for wealthy immortals who care nothing for those lesser beings they control even as they feed off them. They are called vampires.
Monday, August 19, 2013
There was a time, in my innocent youth, when the follow-up note for the average cancer patient (usually scrawled in quasi-legible doctorese) looked something like this:
S: No new complaints.
O: Px-unchanged. Responding on CXR. Labs OK.
A: Doing well
P: Continue Rx. RTC 3 w.
I thought of this recently when, preparatory to starting my new clinic at Stanford, I took Epic training. I had moved from a healthcare system that used Cerner, and of course when you know one electronic health record system, you know one electronic healthcare system. After spending some time getting access (yet another set of passwords and another user name), and getting "tokens" for my iPhone, my iPad, and my Mac, I spent an hour or so with a lovely lady assigned by the hospital to explain the new system to me, who then handed me a helpful booklet that would remind me when I inevitably forgot the lesson. Which I did, almost immediately.
Epic Systems is headquartered in Verona, Wisconsin, a sleepy town on the outskirts of Madison. My dad used to take me there for haircuts when I was a teenager, and so it is fitting that Verona continues to provide me haircuts via Epic. The cuts are now to my time, and my emotional comfort. I’m hoping both will improve as I become more at ease with the system, though I have yet to meet anyone (including hospital administrators) who thinks it will speed me up.
Epic is based on 1960’s MUMPS software developed at Massachusetts General Hospital. Think about that: a software system that pre-dates MS-DOS. It is (and this is not particular to Epic) loaded with excess clicks, small fonts, long scrollable lists of diagnoses in no particular order, and buried data sets. The company was founded by a former employee of the University of Wisconsin’s Psychiatry department, the sort of fact that is almost too delicious: are we all part of some devious, extreme stress-inducing psychological experiment designed to increase the business of the psychiatrists? Well, that would just be paranoid, wouldn’t it? Really, I’m not crazy. Really.
Comparing 1983 with 2013, the 4 lines of meaning mentioned above are buried somewhere in pages of cloned busywork stored somewhere on a server and accessed via a clinic workstation. Has the current electronic health record made us better healthcare providers, or more efficient at getting through the day? That these are still arguable propositions (and they are regularly argued, at least in the precincts I hang out in) says a great deal about our tendency to adopt new technologies without putting them to the test.
Modern healthcare record keeping dates back to the 1920’s, when the American College of Surgeons created the Association of Record Librarians of North America to “elevate the standards of clinical records in hospitals and other medical institutions.” ARLNA still exists, transmogrified to the American Health Information Management Association (motto: Quality Healthcare through Quality Information), with more that 64,000 members. But the records it shepherds are no longer primarily under the control of physicians, nor written on paper. Like the rest of modern society, they have gone digital.
What has driven the digitization of the current medical record is fairly straightforward. First, the need to document, for the benefit of the payers, that we are doing what we say we are doing. This is a function of the American healthcare system's radically dysfunctional payment scheme, with its a la carte menu approach, propagated by government but equally embraced by private insurers.
Second was the vision of the electronic health record as sovereign cure for our all our woes. This is related to, and at least partially dependent on, the documentation mandate, but was clearly something more. Back in 2005 the RAND corporation, one of our premiere think tanks, predicted that the rapid adoption of EHR technology would save the U.S. healthcare system more than $81 billion annually through improved efficiency. The number was not plucked out of the air: it was a reasonable extrapolation based on information technology’s effects on other aspects of the American economy.
Politicians -- both Republicans and Democrats -- were quick to latch on to the latest promise to reduce healthcare costs. George Bush and Barack Obama considered it essential for American physicians to adopt EHR.
Earlier this year Arthur Kellerman and Spencer Jones of the RAND Corporation provided an update in Health Affairs. To summarize the healthcare savings benefits achieved through EHR adoption: NOT! Why not? “The disappointing performance of health IT to date can be largely attributed to several factors: sluggish adoption of health IT systems, coupled with the choice of systems that are neither interoperable nor easy to use; and the failure of health care providers and institutions to reengineer care processes to reap the full benefits of health IT.”
The cost savings were illusory in part because, in contrast to other industries, EHRs are used by hospitals to help game the system, increased documentation leading to higher levels of billing. So while EHR systems are expensive to purchase and maintain, the return on investment for a hospital corporation can be impressive.
I can certainly speak to the “not easy to use” part. There is nothing easy or intuitive about EPIC or Cerner or a their kindred: they look like a Boeing 747 control panel. Actually, let me take that back: I cannot imagine that they are designed like a Boeing 747 control panel, because if they were, then there would be far more fatal airplane crashes than actually occur.
If EHR adoption has not improved physician efficiency or reduced healthcare costs, who has benefitted? EHR vendors lead the list, of course. The big EHR firms (Epic, Cerner, Allscripts) have all seen explosive growth in recent years. This growth is driven by federal government incentives (the recession stimulus package) and federal mandates (the 2015 deadline for meaningful use of EHRs). Both carrot and stick were the result of heavy lobbying efforts by EHR companies, which like the pharmaceutical companies are now an integral part of the Washington “you scratch my back, and I’ll scratch yours” reciprocal favor system. Allscripts’ CEO visited the White House on seven occasions after President Obama took office in 2009, and personally made more that $225,000 in political contributions.
Epic saw its sales double in just four years to $1.2 billion in 2011. Cerner has revenues of $2.2 billion, and Allscripts comes in somewhere in the same range as Epic. EHR vendors never suffered during the economic downturn. These guys have done quite well for themselves, if not always for us.
The basic premise behind EHRs, and the reason no one wants to give up on them quite yet, is still sound: having all the data instantly available at your fingertips should make patient care safer and more effective. If this has not yet happened, it is in part related to lack of interoperability between the EHR systems. Crossing town from one practice or hospital to another is frequently to enter a deep electronic chasm. The best you can hope for is that you live in a place where monopoly prevails. Epic (and I really don’t mean to pick on Epic) has the EHR franchise for Stanford, UCSF, and Kaiser Permanente, among other local concerns, so the Bay area is relatively integrated from an EHR standpoint.
But this is not the case everywhere. And without interoperability (which EHR vendors pretty much all passively or actively oppose) the promise of EHRs will continue to be largely theoretical. Gandhi’s famous response when asked what he thought about Western civilization comes to mind: “It would be a good idea.”
ASCO has devoted significant resources (significant for ASCO if trivial for the EHR vendors) to creating CancerLinQ, its rapid learning healthcare system. My sense, in my brief experience with Epic and my older relationship with Cerner, is that the EHR vendors have not thought through the needs of specialties, perhaps because their principal relationship is to hospital corporations.
Electronic health records are an important component of any rapid learning system, perhaps the central component. CancerLinQ will ultimately need to interact with a whole slew of EHR vendors. Will it be embraced, or strangled? Or just ignored? We’ll see. In the meantime, I will continue to plod through my dictations.
Thursday, July 18, 2013
Not long ago, trying and failing to fall asleep, I suddenly became aware of a surfeit of lights staring at me out of the darkness: the moonlight peaking, Fraunhofer-line like, through the slits between the Venetian blinds; the fluorescent hour and minute hands of my watch, glowing green as they slowly, slowly marked the passing minutes and hours; the pale light slinking under the bedroom door from another room; the glowing red on/off button on the power strip plugged in the wall; and the two pinpoint yellow lights where the power cord mated with my computer. All, all suddenly conspired, or so my sleep-deprived mind perceived, to illuminate my bedroom.
Illuminate is probably not the right word to describe this sensation: the total number of lumens emitted by these light sources barely signified. But the mind plays funny tricks at three in the morning. Most babies, and many adults, find night-lights comforting. There is even a psychological disorder, nyctophobia, characterized by a severe fear of the dark.
I am the opposite of a nyctophobe. I require darkness to sleep, and the darker the better.
During World War II night fighter pilots would sit in a dark room before flying off on a mission, allowing the retina’s rods to become super-sensitive. Rods are responsible for night vision, and dark-adapted vision is optimal after 30 minutes or more of darkness. So sensitive are these dark-adopted rods that, some evidence suggests, under ideal conditions their photoreceptors can be activated by a single photon. My rods had turned minuscule points of light into searchlights, high-intensity beams pointed at my reticular activating system. Restless, I got up and went out on my back porch.
The nights in Palo Alto are generally clear, and one can see a plethora of stars. Not as many as I saw in my youth, lying on my back in a field in central Wisconsin; there are too many city lights for that beautiful brilliance.
I wonder if this is how astronomy, that oldest of the sciences, began: some Mesopotamian insomniac looks up at the night sky and notices that the stars resolve themselves into the false order of constellations, and how they trek through the night sky, and track through the seasons. He uses cutting-edge cuneiform technology to note down his findings, becomes Ur’s or Babylon’s temple priest for his efforts (astrology and astronomy being close kin, and academia then as now dependent on government funding). Science is born. Or not. But it is a pretty thought to think at 3 A.M.
These night lights tell stories. Some I am sure you know, some perhaps are less familiar. Start with the fluorescent watch hands. In the early twentieth century there was a radiation craze, and one fall-out of the public passion for being irradiated was the development of radioluminescent watches whose hands were painted with radium. For the first time in history an insomniac knew exactly how much sleep he was missing, a minor advance in the march of human progress. The watches were very popular in their day.
The painters -- usually women -- used delicate camel hair paintbrushes, which they twisted to a fine point using their lips and tongues. They received, for their efforts, about 27 cents per dial, adjusted for inflation, and were expected to paint 250 dials per day. Here they are at work:
Day after day, year-in and year-out, they painted their lips with radium. And then, predictably -- predictable to us, but those were more innocent, or at least more reckless, times -- they developed osteonecrosis of the jaw, anemia, and oropharyngeal cancers. The company denied responsibility, though its chemists carefully used lead screens, masks, and tongs.
As occupational diseases go, it was unique, and of course one never sees these patients today. They were already an epidemiologic legend when I was a fellow three decades ago. But I have seen osteonecrosis of the jaw, and it is ugly: bones turning to mush, chronic infections, long periods with an oral surgeon, and chronic pain.
Those lethal watches had long-term legal consequences: the public outcry over lawsuits against the U.S Radium Company (the watches’ manufacturer) by the so-called “Radium Girls” ultimately led to the enactment of occupational disease labor laws. The employees themselves received a $10,000 payout and a $600 annual annuity. Radium paint was still in use as late as the 1960’s.
Radium poisoning had a popular culture afterlife. Nothing Sacred, a delightful 1930’s screwball comedy, starred Carol Lombard as a woman allegedly dying of radium poisoning. Allegedly, because radium poisoning was a misdiagnosis by her drunkard physician, allowing the lovely Ms. Lombard’s character to live and fall in love. Hollywood can make even environmental carcinogenesis a matter of sport: her character cheerfully states, “You know, I’m not going to bed until I have convulsions and my teeth start falling out.”
The current fluorescence of the minute and hour hands of my watch does not involve radium: today’s phosphorescent paints are either silver-activated zinc sulfide or doped strontium aluminate. I can not sleep soundly knowing that I am not contributing to the deaths of watchmakers.
But there is another radiation source in my bedroom, and it is associated with one of those night lights: the smoke detector. It uses the radioactive isotope americium-241 to emit alpha particles into an ionization chamber. If smoke particles enter the ionization chamber the ions are less capable of carrying a current, and the current drop sounds an alarm.
The light in my particular smoke detector is powered by a battery, not the americium-241, but it still manages to help keep me awake. Given the isotope’s 432-year half-life, this could be a long night. But at least I don’t have to worry about alpha particles, with their low penetrative power.
The stars, the ultimate night lights, tell so many stories that one hardly knows where to start, or end. The constellations each have their own stories, stories that charmed me as a Boy Scout: I still thrill to see Orion’s belt on a cold, clear winter night. Orion, the hunter, banished to the skies by Gaia for threatening to kill all the animals.
But my favorite story is scientific, not mythological: the story of Olbers’ paradox, named after Heinrich Olbers, a 19th century Dutch astronomer (though Johannes Kepler had, even earlier, considered the problem). Assume that the universe is infinite, and that the stars are evenly distributed throughout. Our line of sight should then, in every direction, be filled with stars. Hence Olbers’ paradox: why is the sky dark at night? Why is night not as bright as the day?
The first approach to a right answer (there is more than one right answer, it turns out) came, not from a scientist, but from a writer: Edgar Allen Poe, he of “Nevermore” and “The Pit and the Pendulum.” Poe reasoned:
“Were the succession of stars endless, then the background of the sky would present us a uniform luminosity, like that displayed by the Galaxy – since there could be absolutely no point, in all that background, at which would not exist a star. The only mode, therefore, in which, under such a state of affairs, we could comprehend the voids which our telescopes find in innumerable directions, would be by supposing the distance of the invisible background so immense that no ray from it has yet been able to reach us at all.”
A good enough answer, I would have thought, but there’s another. The Big Bang, some 13.7 billion years ago, was associated with intense heat and, the astronomers say, light brighter than the sun. But most of that radiation has now redshifted to microwave wavelengths. Have you ever seen a microwave when you pop up some Orville Redenbacher? Me neither: my rods and cones aren’t built for them.
If you want to see a brief, nifty explanation of Olbers’ Paradox, go to this YouTube video: http://www.youtube.com/watch?v=gxJ4M7tyLRE
Olbers’ Paradox is resolved by Edgar Allen Poe and Orville Redenbacher popcorn, which I for one find a deeply satisfying outcome. It would be even more troubling if the sky was bright all night long: I would never sleep.
Thursday, June 13, 2013
Every year the ASCO meeting seems to revolve around a few themes, or sometimes memes, that define where we are at and where we are going as a profession. This year’s meeting was no different. Outgoing President Sandra Swain, in her presidential address, presented the meeting’s dominant theme: ASCO as ISCO. A for American, I for International, a reflection of the global nature, not just of our professional society, but also of our scientific collaborations and our collective public health task.
ASCO, to its credit, is doubling the resources it devotes to international activities. Given the large number of ASCO members hailing from outside the U.S., and the even larger percentage of international attendees at the meeting, this is certainly appropriate. But the larger reason is the growing cancer burden around the world, a burden that we should take on, if only by providing educational and scientific support.
The plenary session both re-enforced “ASCO as ISCO” and provided important nuance. In the United States, cervical cancer belongs to the land of rare cancers: it is not a major public health challenge, given the ubiquity of screening and early intervention. In contrast, it is a major public health issue in large portions of the globe, where inadequate health care systems doom large numbers of women to untimely and miserable deaths.
For this reason if no other it was a delight to see a plenary session lecture emanating from India, where a randomized clinical trial demonstrated that visual inspection of the cervix with acetic acid, performed by a non-physician health care worker, was an effective and inexpensive means of screening. The authors estimated that this approach could save 73,000 lives per year on a worldwide basis.
I loved everything about this study. This approach has a potentially greater public health impact than anything I have seen presented at the annual meeting for years. The intervention is low-tech, cheap, and widely available. The trial was funded by our National Cancer Institute: if ever federal government spending for clinical trials needed justification (and, alas, it does, even within the NCI), this trial should provide validation for decades to come.
But the plenary session was not finished with cervical cancer. A separate randomized Phase III trial (also NCI-funded) examined the role of the antiangiogenic agent bevacizumab in advanced cervical cancer, and demonstrated a 3.7 month overall survival advantage. It was, as both the presenter and the discussant pointed out, the first ray of hope for metastatic cervical cancer in many a year.
There was just one false note in this presentation, or so it seemed to me. The presenter, in his conclusion, opined on the need to find a way of making this treatment approach available on a worldwide basis. This remark was clearly well intentioned, attempting to address the disparity between high-income healthcare systems and the many low- or middle-income countries where modern cancer care is unavailable. ASCO as ISCO, right?
And yet, as I have said, it came across as wrong-footed. If vinegar can save 73,000 lives per year at trivial cost, why on earth would an impoverished health care system consider paying for a pricy biologic, requiring highly trained health care professionals, and curing no one?
I know I wasn’t the only one to be struck by the apparent disconnect between these two plenary lectures, and by the stark contrast between the price: life-years saved of the two interventions. Several of my colleagues mentioned it to me as well. The cervical cancer theme became an ASCO meme, or a collection of related memes: the need for early as opposed to late interventions, the value of population-based public health measures as opposed to individual treatments, and the need for us to take a serious look at which therapeutic interventions provide the most bang for the buck.
Add to the vinegar story (a low-tech solution) the success of HPV vaccines in preventing cervical cancer, and the story expands a step further. Prevention, added to early detection, should be our goal for cervical cancer.
Should it be the goal for every cancer? That is the basis for a large number of arguments, as the screening mammography quagmire suggests. Early detection isn’t always cost-effective, doesn’t always save more lives than advanced treatment, and doesn’t always have solid scientific underpinnings. Prevention techniques require huge trials with long follow-up, and even if the intervention is successful (tamoxifen in breast cancer) it may not be used.
But for cervical cancer there is no question: early detection and prevention work. They save lives, and do so at an acceptable price. Late therapeutic interventions (chemotherapy, and now antiangiogenic therapy) cost more, require higher levels of technology and health care personnel, and don’t cure anyone.
Again, I don’t mean this to sound like an attack on the bevacizumab trial. This trial was a success. It is a randomized controlled trial with a statistically significant improvement in overall survival, in a disease where nothing positive has happened for a very long time. Most patients would happily take an agent that extended their life by several months, particularly an agent that is not particularly toxic.
And certainly it is a false dichotomy to say that we should save lives through prevention or early detection, but not prolong the lives of patients with advanced disease. I want us to do both, and certainly as a clinical trialist I would have been delighted to present a positive Phase III trial at my society’s premiere event.
But the two trials showed the difference between what medical oncologists consider important (introducing new targeted therapies that improve outcome for patients with advanced disease) and what rational health care systems consider cost-effective and valuable. The difference between the two is rarely thrust in our faces as it was by the plenary session, but it is always there in the background.
Should we spend money on smoking cessation, on CAT scans for smokers, or for therapy for Stage IV adenocarcinoma of the lung with the latest targeted ALK inhibitor? All of these approaches “work,” depending on what you mean by “work,” but they all mean something different, and all require something different of the health care system in terms of access, expense, and ultimate value. The answer different systems will come to will depend on economic and political calculations as much as medical and scientific formulations
The other emerging theme at ASCO involved immunotherapy. Here, as two years ago, melanoma led the way. If you had told me, a few years ago, that I would soon attend an ASCO meeting where the breast cancer researchers were bored to tears, while the melanoma docs walked around with goofy smiles on their faces as new drugs rained down on them from the Pharma heavens, I would have thought you nuts. But this has come to pass, not once but twice in the past three years.
The star here was anti-PD-1 therapy. Tony Ribas of UCLA presented the results of a study with lambrolizumab in advanced melanoma, and showed an overall response rate of 38%, with many durable responses. Jedd Wolchok of Memorial Sloan-Kettering presented a combination of the anti-PD-1 agent nivolumab with ipilimumab, and demonstrated responses in more than half of treated patients. The anti-PD-1 drugs look like stars: less toxic than ipilimumab, combinable with other agents, employing a novel target, and exportable to other disease types.
Cancer immunotherapy has finally arrived, after decades of false starts. Future ASCO meetings will see a flood of Phase III trials with these agents, alone or in combination with other immunotherapeutics and targeted therapies. One gets the real sense from these studies that melanoma has turned the corner and headed off into a new and quite promising direction. But even more, immunotherapy is here to stay, and oncologists will need to dust off their old textbooks and learn the mysteries of CTLA-4, PD-1, PDL-1, Tregs, and other terms we have forgotten in the years since medical school. Or, in my case, learn for the first time, since these pathways were undiscovered when I was in medical school. But I age myself.
As always I had two feelings about ASCO: anticipatory exhaustion as my plane landed in Chicago, and delight at learning new things and seeing old friends. I routinely feel, on leaving the meeting, that I belong to a vibrant community of dedicated researchers and physicians. That sense of community, that belief in progress, brings me back year after year.
Monday, May 20, 2013
One of my favorite pictures of Winston Churchill shows him, in profile, surveying the ruins of the House of Commons. That storied building, the “Mother of Parliaments,” had been destroyed in the last great German air raid on London in 1941. Churchill was there to survey and pay tribute to the building that had nurtured his political career. The building was, in a very real sense, what the Western allies were fighting for: a place where political discourse, not a dictator’s whim, decided human events.
The House of Commons was dominated by two design elements. First, it was an oblong where the parties literally faced each other. Second, it was cramped, with insufficient room for all the elected parliamentarians, and with no desks for individual members.
The Nazi bombs offered a choice: recreate the old, or build something “modern,” less cramped, more “functional,” with all the modern amenities. Churchill never seriously considered the second option. There was, he thought, magic in the old structure, with the seating arrangements forcing one to address one’s political foes, the tight space creating excitement and political conversation. Anyone who has ever seen debates in the House of Commons, and then contrasted them with the empty spaces (I refer to the building, not the politicians) one sees on C-Span will appreciate Churchill’s analysis.
As he told Parliament (in a statement much beloved by architects), “We shape our buildings, and then our buildings shape us.”
It is a quote I often consider when passing by buildings great and small. Buildings send messages, often obvious ones: a Chartres cathedral, aspiring to the heavens; a Parthenon, simultaneously glorifying human reason and the free citizens of Athens; or the massive Stalinist architecture of the 1940’s and 50’s, as ugly in form as the tyranny it represented, with its “your individuality means nothing to the State” message.
At a somewhat lesser level, consider courtrooms. Hollywood courtrooms are often glorious, but their current counterparts, of the sort one sees regularly on second-rate reality TV, are as sordid as the crimes. The majesty of the law seems somehow less majestic in low-ceilinged, fluorescent-lit, faux-wood paneled modern courts. I’m sure they are less costly to build, but does anyone really care about these structures?
Is this worship of form just romantic nonsense? Or do our buildings really shape us? The question is pertinent to science and medicine. Consider research facilities. If one believes that one can design an environment that fosters creativity, then why would we not want do so? If proper building design improved research productivity by 20%, then the indirect costs from research grants would pay for the cost of just about any structure in short order.
Of course, research buildings don’t exist just to foster creativity. They are also there to carry out the physical act of experimentation, and that physical act imposes real design constraints. Creativity and experimentation are two different, if related goals: consider Princeton’s Institute for Advanced Study and the Lawrence-Livermore laboratories and you get some idea of what I am talking about. But let’s focus on the creativity part, on the assumption that you can always find a structure to actualize a great idea.
Steve Jobs famously believed in the mysterious alchemy created through building design, and both Pixar and Apple headquarters are exemplars of this concept.
The Pixar campus has a great atrium space at its center, acting as a central hub for unplanned encounters. The atrium houses mailboxes, cafeterias, workout rooms, and restrooms. It forces employees to come together. And it is attractive, with Jobs’ usual concern for aesthetic design: a place you want to work, rather than one you have to. It also has sculptures recounting Pixar’s successful movies, reminding employees why they work .
Bear in mind that Steve Jobs did not believe in the mystical powers of distributed networks, but rather in the socializing values created through proximity and the personal touch. Jobs, like Churchill, believed that we shape our buildings, and afterwards they shape us.
But is it true? The House of Commons and Pixar are pretty good anecdotal evidence supporting the hypothesis, but in the end just anecdotes. The hypothesis is, I suppose, testable. Take two groups of equally well-trained knowledge workers and randomly allocate them to a “control” building designed by low budget bean counters, or to a Jobs-type building environment. Come back five years later and measure outputs (patents, animated movies, scientific citations, or whatever) and compare. No one, to my knowledge, has conducted the experiment.
Instead we have what might be considered, if architects were oncologists, a series of underpowered Phase II trials of buildings. And a fair number of them, including the Salk Institute and Stanford’s Clark Center, as well as numerous attempts scattered through many industries.
It is difficult to imagine a scenario in which putting mediocre investigators into great buildings will turn them into great investigators. Similarly, if you tore down Pixar’s campus and put its talented animators into a warehouse in downtown Fresno, they would still be likely to outperform. Great scientists would still publish in Nature or the New England Journal of Medicine even when placed in a mediocre building (most do already, after all). Social engineering (of which the “our buildings shape us” argument is a subset) has very real limits. Leadership (which creates a different sort of environment), resources and talent all matter.
Nevertheless, it is equally hard to imagine that we couldn’t do better, couldn’t be more creative, if we fostered interaction, and one way of fostering interaction clearly involves one’s physical environment. Unfortunately, because space is always at a premium, the “touchy-feely” aspects embodied by the Pixar atrium are often jettisoned in our unquenchable thirst for bench space and maximal office footprint. The standard way universities allocate space (X dollars of indirect costs = Y dollars of square footage) actively undercuts this sort of intellectual agora.
Buildings can shape us in other ways. I have spent a fair amount of time at the National Academy of Sciences’ Keck Building in Washington, DC doing committee work the past two years, and have come to be charmed by the place. The building itself is not beautiful, nor even particularly functional, but it has a lobby that others could well emulate. Carved into the marble are emblems of scientific success: formulas, engravings of Darwin’s finches, the original X-ray crystallography for DNA.
Every time you walk through the door, you are reminded of the great scientists who helped create the modern world. Does that matter to the people who work at Keck? I don’t have a clue. But it is wonderful nevertheless. And I hope someone will help me with the equation.
Hospitals appear to be a special case. They are, by necessity, utilitarian. They are usually built by accretion over time, with little logic. Can they be designed to foster high-class medical care? There are certainly design firms that think so.
Bad design certainly can cause massive problems. I knew a hospital where the architects, new at the job, made the doors to patient rooms quite narrow, forgetting that hospital rooms have full-bodied hospital beds. At enormous cost the hospital special-ordered beds that would fit through the doors. Sometimes we shape our buildings and they drive us crazy. But how about good design?
As with research buildings, hospitals (and clinics) have purpose. That purpose, simply put, is to take in sick people and send out healthy people. That purpose, by general consent, has two sub-purposes: healthcare should be skillful (i.e., it should deliver effective care efficiently) and it should be compassionate (i.e., it should emulate the great tradition described in the Oath of Hippocrates).
Can we shape hospitals in ways that shape us to be skillful and compassionate?
I don’t know. I have certainly seen designs that are more or less efficient, depending on where one places the ER or the radiology suites or the lab. But efficiency is only one aspect of skill, and it need not be the least bit compassionate, other than in the sense that a short length of stay exposes patients to fewer nosocomial infections and less hospital food.
A solitary focus on efficiency may even thwart the development of caring teams. Teamwork and creative care are the provenance, not of the building, but of the employees who walk the halls.
There’s a wonderful sight gag in the Marx brothers’ A Night in Casablanca, where Harpo (the one who never speaks) is leaning against a wall (see the clip on YouTube here). An officious busybody accosts him. “Say, what do you think you’re doing, holding up the building?” Harpo happily nods yes. “Come on!” says the official, pulling Harpo out of the way. The wall promptly falls down. We shape our buildings, and then our buildings shape us, but like Harpo it is we that keep the walls from falling down.
Data is temporarily unavailable. Please try again soon.