I was listening to the journalist Ryan Grim of Drop Site News last week as he talked briefly about his wife’s ongoing cancer treatment. His short gloss was directly on point, and it motivated me to dig deep in my archives on the history of research on breast cancer and how one never knows what scientific knowledge is “useful” until it is. [1] Grim’s message, that current cuts enacted across the board will severely damage health in the United States, is especially important for our time.
Cancer has been understood as the unregulated growth of tumors and their spread (metastasis) for a long time. Theodor Boveri wrote Concerning the Origin of Malignant Tumors in 1914:
Theodor Boveri’s Zur Frage der Entstehung maligner Tumoren has had a momentous impact on cancer research. In it he argues that malignancy arises as a consequence of chromosomal abnormalities and that multiplication is an inherent property of cells. With astonishing prescience, Boveri predicts in this monograph the existence of tumor suppressor mechanisms and is perhaps the first to suggest that hereditary factors (genes) are linearly arranged along chromosomes.
If you find the CHSL Press edition in your library or for a few dollars at the used bookstore, you won’t be disappointed. Nevertheless, until the 1970s and 1980s cancer was largely a molecular mystery. Now we know that the first step in tumor progression is usually the loss of control of cell division that produces the primary tumor. Subsequent mutations then allow the cancer cell to move to other places in the body and form secondary tumors (metastases). These are hard to treat and are usually what kills the patient, slowly or rapidly depending on the cancer. In my reading, the best single source on cancer as a disease remains The Emperor of All Maladies: A Biography of Cancer (2010) by Siddhartha Mukherjee.
Although this is not generally understood, the molecular bases of cancer were established in research that did not focus directly on cancer. Rather, biological scientists’ goals were to understand fundamental processes such as the control of cell division and cell proliferation, the control of gene expression, the regulation of metabolism, and mechanisms of cell motility. Most Cancer-causing oncogenes were originally associated with viruses that can cause cancer. Upon further research it was discovered that viral oncogenes are normal genes that were taken up by the virus and then expressed in infected cells, thereby causing cancer. The normal gene of a cognate oncogene may be called a proto-oncogene, but the “proto” is often left off in current usage. Mutations in these genes, separate from infection by viruses carrying an oncogene, lead to cancer when the mutant protein is expressed.
Oncogenes have become a subject in their own right, and targeting the function of oncogenes has become a very productive focus of clinical oncology. HER-2/neu is the specific example for today. But before we get to that, the basic research must come first. In 1962, Stanley Cohen of Vanderbilt University published “a deceptively arcane study on the isolation of a protein that could accelerate incisor eruption and eyelid opening in newborn mice.” The original name for the protein was “tooth-lid factor.” The protein was eventually named “epidermal growth factor” (EGF) and led Cohen to develop a new area of cell biology for which he and Rita Levi-Montalcini, who discovered a similar protein she named “nerve growth factor” (NGF), were awarded the Nobel Prize in Physiology or Medicine in 1986.
EGF and other growth factors function by binding to their specific receptor on the target cell and direct the cell to proliferate through a series of evolutionarily conserved mechanisms inside the cell. For our purposes: EGF + EGFR Active EGFR growth. When this pathway remains on when it should be off, a result is cancer. Several EGF genes are encoded in the human genome. The first human EGFR was called HER-1, for human EGFR-1. Similarly, the current name of EGFR-2 is HER-2 (also called neu because it was also identified in brain cancer). When the gene for HER-2 was sequenced it was found to be essentially the same as v-ErbB (i.e., viral oncogene B from avian erythroblastosis virus (AEV), a retrovirus that causes leukemia in chickens). Thus, v-ErbB is the viral oncogene that leads to cancer in chickens when they are infected with avian erythroblastosis virus (AEV) that previously “grabbed” the chicken EGF2 gene during an infection. HER-2 is the homologous proto-oncogene that causes cancer in humans when its expression is altered.
The importance of HER-2 for cancer in humans was recognized in the late 1980s when it was found that HER-2 is strongly associated with breast and ovarian cancer. I presented the paper Studies of the HER-2/neu Proto-oncogene in Human Breast and Ovarian Cancer in a departmental seminar shortly after it was published in 1989 and have followed this research ever since. It is a remarkable story. Other research groups have done much research on HER-2, but as I remind my medical students, this work was done by individual scientists before it appeared in their textbooks. Dennis J. Slamon of UCLA has been a leader in the field. From the Abstract:
Carcinoma of the breast and ovary account for one-third of all cancers occurring in women and together are responsible for approximately one-quarter of cancer-related deaths in females. The HER-2/neu proto-oncogene is amplified in 25 to 30 percent of human primary breast cancers and this alteration is associated with disease behavior. In this report, several similarities were found in the biology of HER-2/neu in breast and ovarian cancer, including a similar incidence of amplification, a direct correlation between amplification and over-expression, evidence of tumors in which overexpression occurs without amplification, and the association between gene alteration and clinical outcome…The data presented further support the concept that the HER-2/neu gene may be involved in the pathogenesis of some human cancers. (emphasis added)
At the time, this made perfect sense:
- HER-2 is an EGF receptor on the surface of breast and ovary cells, among others, where it controls proliferation of these cells.
- When HER-2 is amplified the cells are induced to proliferate in an unregulated manner.
- Dysregulated cell growth/proliferation leads to cancer.
The identification of the oncogene v-ErbB with HER-2 led directly to a potential therapy for breast cancer that has worked very well. It took nearly twenty years, but Herceptin, i.e., the monoclonal antibody trastuzumab that targets HER-2 on the cell surface, was one of the first targeted chemotherapies. Prior to the advent of targeted chemotherapy, the goal of chemotherapy was to kill cancer cells faster than normal cells. This works well now for many cancers, but targeted therapy has revolutionized clinical oncology.
Herceptin (Genentech) has been a lifesaver for many cancer patients, and this happened as a direct consequence of basic research on something as outwardly unremarkable as tooth eruption and eyelid opening in neonatal mice. This research was, of course, funded by the National Institutes of Health and other agencies. A few key points include:
- The oncogenic potential of HER-2 was indicated by its identity with v-ErbB, an oncogene in a chicken retrovirus (which is where the first oncogene identified from Rous Sarcoma Virus (RSV). What happens, or doesn’t, in one vertebrate probably also applies to another. [2] As noted here before, this apparently has been forgotten regarding the problematic concept of durable immunity to coronaviruses such as Infectious Bronchitis Virus (IBV) in chickens and SARS-CoV-2 in humans.
- The function of the EGF receptor (HER-1, HER-2): Without going into molecular detail, EGFR must dimerize to transmit its signal into the cell and tell the cell to grow. This requirement was determined in research over many years in many laboratories. Herceptin does not prevent soluble EGF from binding to its receptor but it does interfere with dimerization, and therefore the signal to grow is disrupted. Another antibody developed later, Pertuzamab, prevents HER-2 from dimerizing with its preferred partner HER-3 and works in a complementary manner.
- Thoroughly basic biomedical research on signal transduction through EGFR-like molecules revealed that receptor signaling is downregulated when the cell engulfs the active receptor, thereby removing it from the cell surface. When a conjugate of Herceptin and the cytotoxic drug emtansine is internalized, the drug becomes a “magic bullet” that kills the cancer cell after preventing it from dividing. Two hits are better than one.
- Based on studies of many cancers, Herceptin can be used to treat several HER2-positive cancers, including metastatic breast cancer and gastric, lung, and colorectal cancer.
Herceptin/trastuzumab is on the World Health Organization List of Essential Medicines, as it should be.
The story of HER-2, breast cancer, and Herceptin illustrates the truth that basic scientific research, if it is to be effective, cannot be prescribed, whatever the current leadership of American science seems to think. The corollary is that basic research leads to clinical relevance in unexpected ways. Would Herceptin have been developed if Stanley Cohen had not discovered what was eventually known as EGF, epidermal growth factor? Probably, but perhaps much later than necessary, for millions of cancer patients and their families. But maybe not. Or perhaps the trajectory of the alternative path might have pointed in a different direction without the same significance.
Would Biotech and Big Pharma have developed Herceptin without the initial research on HER-2/neu? Certainly not, for two reasons: (1) In the first place, the basic knowledge would not have existed, and (2) Big Pharma depends on publicly funded science to provide the foundation for their production and marketing of virtually all their products.
The current attack on basic science during Trump v2.0 has garnered some sympathy, primarily because our scientific leadership has lost the plot at the very visible margin. Having been an insider of sorts for more than forty years, I have seen the good and the bad of science funding at NIH and NSF. But the solution is not to tear down the edifice.
One related example from the case of HER-2 and breast cancer stands out. The original reports of the connection were controversial. This was dispelled in one of the most important follow-up papers, Detection and quantitation of HER-2/neu gene amplification in human breast cancer archival material using fluorescence in situ hybridization (1996). From the Abstract:
Amplification and overexpression of the HER-2/neu gene occurs in 25-30% of human breast cancers. This genetic alteration is associated with a poor clinical prognosis in women with either node negative or node positive breast cancers. The initial studies testing this association were somewhat controversial and this controversy was due in large part to significant heterogeneity in both the methods and/or reagents used in testing archival material for the presence of the alteration…Fluorescence in situ hybridization (FISH) represents the newest methodologic approach for testing for this genetic alteration. In this study, FISH is compared to Southern, Northern and Western blot analyses as well as immunohistochemistry in a large cohort of archival human breast cancer specimens. FISH was found to be superior to all other methodologies tested in assessing formalin fixed, paraffin embedded material for HER-2/neu amplification. The results from this study also confirm that overexpression of HER-2/neu rarely occurs in the absence of gene amplification in breast cancer (approximately 3% of cases). This method of analysis is rapid, reproducible and extremely reliable in detecting presence of HER-2/neu gene amplification and should have clinical utility.
It most certainly did have “clinical utility.” And the key to this success is something very prosaic that has nevertheless become a lightning rod: Indirect costs. Without adequate institutional support that is funded by indirect costs, the large cohort of human breast cancer specimens essential to confirming the HER-2/breast cancer connection would probably not have been available. These were saved in the tissue archive because indirect costs/overhead made this possible. Every biomedical research lab as sample archive of some kind and much effort goes into maintaining this, which is large laboratories can be a fulltime job.
Biological samples are often stored in cabinets and/or ultracold freezers that are expensive to buy and maintain. This cannot be done using the current research budget, but current research is always dependent on the foundation that includes samples from previous research. These samples also contribute to research in other laboratories across the world. When I directed my own laboratory, requests for samples were a constant thing, and they were always granted. Could the HER-amplification hypothesis have been confirmed without this archive? Yes, but with a delay measured in years and increased morbidity and mortality due to breast cancer in which HER-2 is amplified.
The National Science Foundation has implemented a standard 15% indirect cost rate effective May 5, 2025. Previous rates were in the 45-55% range in my experience. One of the justifications for this cut is that private funding agencies restrict indirect costs to 10-15%, so why is the government paying more?. This is true. It is also true that, as I have said too many times lately, in the past private funding agencies such as the American Heart Association and American Cancer Society have not approved grants unless the applicant could show evidence of substantial other support for the research. This support would be in the form of grants from NIH or NSF or institutional start-up support for a new academic scientist (soon to be extinct). AHA/ACS awards are icing on the cake, not the cake. If this change at NSF is permanent, research institutions will necessarily reduce support and direct research costs will then make up the difference to the extent this is possible. Research output will decline precipitously. That seems to be the goal.
The ramifications of this reversal cannot be predicted but they are predictable. The connections between v-ErbB, EGF, HER-2, and breast cancer were an “unknown unknown” sixty years ago. Similar connections will be missed as the American research community declines in numbers and strength and reach. American science can be improved, however, if that is a goal. One example can be found in the career of Stanley Cohen, who discovered EGF. He maintained a small laboratory and was immersed in the actual work. The results were spectacular.
Most of the scientists who taught me did the same (albeit without the Nobel Prize). Studies have demonstrated that diminishing returns, such as they can be measured by publications, set in with the second NIH grant (very few scientists receive more than on NSF grant at a time). This has been my experience observing labs that grow for the sake of growth. Spreading the wealth, so to speak, would work, and decentralization would lessen the influence of an erstwhile “leadership” with other ideas. In any case, fifty percent cuts in total support with the remainder devoted to Administration priorities, whatever they are, will be the death of American science as it is practiced while leaving no prospect for rebuilding an American science as it should be practiced.
Notes
[1] As always in this series, following the philosopher Nancy Cartwright, the goal of scientific research is not to produce “truth.” The goal is to produce useful knowledge in an open-ended fashion. Thus, old or unrelated results are often useful in unexpected contexts, such as the connection between viral oncogenes and normal genes that are mutated in cancer as discussed here.
[2] Theodosius Dobzhansky: Nothing in biology makes sense except in the light of evolution. The paper at the link is a rabbit hole I never expected to encounter. But in my view the meaning of this most famous quotation is this: Biological structure and function are explained by biological evolution. For example, my research has shown that the basic structural components responsible for eukaryotic cell motility are more than 1.7 billion years old. Humans have several proteins essential for cell motility that are found in the extant representative of the earliest branch on the evolutionary bush that includes humans. That cancer in birds and mammals is caused by similar mechanisms follows if the former is true.
15 comments
Comments are closed.
Thank you, KLG.
> If this change at NSF is permanent, research institutions will necessarily reduce support and direct research costs will then make up the difference to the extent this is possible. Research output will decline precipitously. That seems to be the goal.
We seem to be on very short notice doing, with respect to the overall US research enterprise, what was done over decades with respect to domestic manufacturing.
One suspects that rest-of-world, especially China, will not hesitate to take advantage of a reverse brain drain from US.
In future, perhaps US economy will be primarily ag and extraction. along with some residual services.
This news item seems apposite.
In addition to international trade war, perhaps there will be an international bidding war for researchers looking for more stable institutional arrangements.
……. “But the solution is not to tear down the edifice.”
Then what is the (or even just one) solution? Should we continue to fund Ralph Baric’s R&D? I agree that the cuts happening now are draconian, but the science of the last few decades has produced some very draconian results. Covid is the result of scientific research as well as the MRNA vaccines (tested without a placebo group). Some edifices need to be torn down, many of them. How do we focus the tear down? We must move beyond complaining.
I think the underlying premise of the cuts, that the current funding arrangements are “too expensive”, is flawed. The USG is a monetary sovereign — it creates the currency that it spends — and is not budget-constrained. There are real world capacity and resource constraints that are the true limits on what USG can accomplish through its fiscal operations. I think the US is nowhere near the limits its potential for research activities. The current policy seems certain to reduce existing capacity in the near term, and it may take a long time to rebuild if it is later decided that the policy was unwise.
IIRC, KLG and others have argued that the current funding and publications systems are sub-optimal. I don’t recall details, but I think that there have been suggestions for improving these.
Personally, I would like to see something like a “Job Guarantee” for STEM graduates. One set of useful activities to which a massive Federal research force could be applied is replication of published results in a much enlarged ecosystem of Federal laboratories.
It has been pointed out that high-risk (i.e., of uncertain outcome) fundamental research is unappealing to for-profit enterprises, which are cash-flow constrained in ways that monetary sovereign governments are not. States, and especially monetary sovereign states, are in a much better position to fund this. Back to MMT, the idea that “USG must cut back on its research expenditures for the sake of fiscal responsibility” is objectively mistaken and I think will in long term prove to be self-defeating.
> It has been pointed out that high-risk
The name escaped me at original writing — a feature of middle age, I think — : Mariana Mazzucato has written about the importance of government investment as a driver of innovation.
Here’s an older NC item on the subject.
Possible but unproven. The failure to properly contain covid was the result of politics and misleadership, which exist independently of the “lab leak or zoonosis” question.
Nonsensical statement
? Straightforwardly untrue
Thanks for this post, KLG. In reading it I was struck by how difficult this work is and trying to put together different pieces to the problem of cancer. It’s kinda like trying to assemble a huge jig-saw puzzle in the dark and not knowing what the eventual picture is supposed to look like until the puzzle is completed when only then will the lights go on.
My closest friend has a case of breast cancer running in her family. Her mother died young, her grandfather, her cousins have it. 25 years ago there were first signs. Then it disappeared. 10 years ago then it resurfaced as “triple negative” diagnosis. I was there when doc told her. I suggested doc – who was an excellent doctor – but bad in talking to people, she might wanna change her lingo, because to any common patient “triple negative” sounds like a death sentence.
However the delay of 15 years might have saved her life as therapy had made such a progress in those 15 years. On the other hand among the 7 or so women who formed a self-help group 2 died. It´s sad and its uplifting at the same time looking into how this has changed and developed.
Btw there is an overlooked little film about the history of how BRCA1 hereditary breast cancer gene was slowly accepted as a medical fact: “Decoding Annie Parker” (2013) with Samantha Morton (known from “Minority Report”) and Helen Hunt as the scientist Mary Claire King who had to fight to be taken seriously with her hypothesis.
100 min.
https://ok.ru/video/340818987648
https://en.wikipedia.org/wiki/Decoding_Annie_Parker
This is not necessarily a film for Sunday´s Coffee Break and it has its weaknesses but since it fits the subject.
Medical topics are difficult to cover with movie fiction.
And thank you very much for doing the work for this entry.
Thanks KLG, and what happens when someone can’t pay to keep the freezers going.
I imagine petal with her lab experience can relate to this sad tale…
What I gather is that there are three (often intertwined) activities — RDA:
1) Research: figuring out how the world functions, explaining some observable phenomena, understanding what some elements are for.
The article details the story of understanding how teeth and lid growth occurs, what are the genes and biological components involved, what are their roles, and how they can further interact to engender cancer.
2) Development: elaborating a solution to a problem.
Again, the article gives the example of cancer therapy based on molecules preventing the EGF growth factor to bind on receptors, respectively preventing EGF to dimerize.
3) Application: turning the result of development, i.e. a laboratory result, into a scalable, useful solution.
That is what herceptin and trastuzumab are. For medical products, that is where those clinical testing phases (I, II, III) come, as well as galenic (pills or solution? Can they be produced in large quantities — will the active ingredient mix well? What is the shelf life of the result?) Same issue for electronics: the lab solution may look impressive, but if the yield on actual production lines is too low, then a product based on that development will not be viable.
In times past (AT&T, IBM) corporations did a tiny bit of Research, and a fair amount of Development. Nowadays, I have the impression that they have completely given up the pretense of doing Research, considerably reduced Development and are focussing on Application. Which also means that DOGE squeezing the pipeline upstream is going to be fatal already in the medium term.
Regarding those sample archives:
“Every biomedical research lab as sample archive of some kind and much effort goes into maintaining this, which is large laboratories can be a fulltime job.
[…]
current research is always dependent on the foundation that includes samples from previous research. These samples also contribute to research in other laboratories across the world.”
Let me point out that archives are essential in other disciplines as well. The links section contains a reference to an article where paleontologists figured out a revision in the tree of life based on (re)studying a 30-years bone sample. I have read countless articles describing such advances based on decades-old samples stored in museum vaults. When trying to figure out the genealogy of AIDS, blood samples kept in medical archives since colonial times proved essential. Climate research made use of meteorological data from century or even centuries-old ship logbooks kept in libraries.
Archives, museums, libraries: all this is threatened by those neoliberal, State-cost-reduction, and DOGE fanatics — and of course undermining the basis for scientific research.
Thank you KLG. Clearly cancer research has become very sophisticated. But finance has not. The term “costs” is misapplied in med research, and absurd. For the usual nuts and bolts development of the world, investments in blunt and obvious solutions pays off, usually by cutting costs by exploiting people and natural resources and that payoff is considered the property of the investors to reinvest however they like in whatever they can get away with. But “costs” in medical research are what it will cost humanity in the future if we do not do the research. We want to prevent those costs altogether. We can only do that by funding research in a timely, long term manner. Long term because that research evolves and is interdependent. It doesn’t fit the proverbial balance sheet at all. So why do we try to make it fit the demands of archaic (and evermore disastrous) finance?
I remember going over to Stanley Cohen’s house with my parents when I was growing up in Nashville. This was probably early 90s and we asked to see his Nobel prize medal. He had to dig through a couple of desk drawers before he found the one he had stuffed it in haphazardly. I knew his research was something about growing skin, but it’s great to learn that it has had such a vital legacy.
I don’t disagree. My argument is against using gain-of-function research as a beard to hide dangerous biowarfare research as something that’s positive for human health. Let us call a spade a spade, as they say. / my 2 cents
Irony. Once you have some but not unequivocal evidence of effectiveness, we’ve had tools for this for decades. Statistical design.
Latin Squares
Balanced Incomplete Block Designs
Etc
One of the problems is that the Stanley Cohen model of research (small lab, focused projects and mentorship, maintaining working knowledge of experimental processes in the lab) is dead, for all intents and purposes.
I worked for a respected PI in graduate school with a lab of nearly 20 students and post docs. He was a complete workaholic with no family or life to speak of outside of work. He probably worked 16-18 hrs most days, and was still in his intellectual prime (late 40s, early 50s) when I worked with him.
Despite the time he applied, his lab was too big for him to track what was going on. The low-performing lab members were essentially completely on their own. He was excellent at grant writing, and our lab was flush with cash and we churned out high impact publications regularly.
The amount of waste and mismanagement of funds I observed would be completely unconscionable in any other profession. There were literally tens of thousands of dollars of reagents that expired, everyone ordered their own material and there is no mechanism for sharing between the labs.
At year end, the department put together a report on per-person per lab costs. We were absolutely by far the leader at a multiple of 2-3x the next nearest lab. He spoke about this almost proudly, claiming that our expenses were high because of our work, and that the costs didn’t matter because we would just get more money anyway in the next grant cycle.
Now, remember, this was a good, productive lab scientifically, if the only benchmark was research output and prestige. But the wastefulness and largesse is utterly out of control and I know that my experience is not unique.
The purpose of this story is not to justify the cuts, but I’d like to also point out that the circuitous discoveries cited in this article arose serendipitously via discoveries made by multiple independent researchers. This is not strong evidence that the system we have works well; should these discoveries be made as part of a single, planned, long-term research project, we can give the “system” it’s due. But I do not believe this “system” deserves credit for the discoveries. The ingenuity of the investigators, sure. But we need to be honest about the shortcomings and largesse of the research enterprise where professors sit atop like feudal lords over the graduate students and post docs (peasantry) and make zero efforts to reform a system that we all know needs it badly.