Let us once more view terrorists or rogue states as “cancers” of the world –for sake of the pedagogical effect of metaphors. Politicians and strategy experts think that this mental picture can teach us how to fight expanding and evasive enemies. But I will argue that the lesson goes rather in the opposite direction: That stealing a page from security politics and the military may change our perspective on how to treat (or how not to treat) cancer in patients. But why bother since after all, we scientists and doctors are so much smarter than politicians? Not so fast. Political sciences and warfare theory are much ahead of cancer research when it comes to the intellectual climate needed for dealing with a major elusive threat that requires an integrative approach beyond use of force.

How killing and sanctioning can backfire

President Barack Obama’s détente with Cuba has triggered the expected kneejerk reflex among conservatives: Why reward Fidel Castro without asking for concessions in return? Hardliners and hawks demand continuation of the existing trade embargo until Castro breaks. The answer of progressives is that it has not worked in the past 50 years, so let’s try something new. Well, at least there is a debate going on. Of course, the proverbial definition of insanity is still: to keep doing the same thing and expect different results each time.

The same argumentation dynamics is more pronounced in the war on terror: Hardliners embrace the straightforward rationale that we should invade enemy countries that harbor terrorists, find them and eliminate them all, because after all, they are a metastasizing cancer that must be stopped from spreading. By contrast, the counter-argument by the more thoughtful experts, as is by now well-appreciated, is that the use of force against the terrorists would only magnify the problem: it would stimulate their recruitment and strengthen their resolve. An attack would make them stronger not weaker. The same with dictators: More sanctions will give them excuses and allow them to consolidate their position. Instead, sometimes a “softer”, “holistic” approach that requires understanding the cultural and economic conditions that breed terrorism and stabilizes bad regimes, and avoiding provocation, may in the long run be more successful.

From the Center for Army Lessons Learned (CALL)

Thus, there are two lessons that cancer researchers, drug companies and oncologists could learn from how our society deals with geopolitical threats but unfortunately have not learned:

LESSON 1 — First, if something has not worked for a long time, we must consider the very possibility that there exist strategies beyond “staying the course” (“the goal is to eradicate all terrorists”) or even “doubling down” (“we need more troops”) even if an existing approach is mechanistically plausible (“kill the bad guys”). We must have the discipline to overcome the mutual reinforcement of ideas, and instead, engage in mutual critique. We must be brutally open to admitting failures and analyzing them and embracing counterintuitive solutions that lie outside our intellectual comfort zone. This will enable collective introspection and thinking outside-the-box of current dogmas cemented by groupthink. All this is not new in military and national security politics.(Such self-assessment and self-correction is even institutionalized in the military). But there is only one doctrine in cancer treatment and nothing outside of it: kill the cancer cells.

LESSON 2 — The second lesson pertains to the specific mechanisms of our failure in fighting a resilient, adaptive and proliferative enemy: our very action may backfire, that is, make the situation worse. This dynamics is precisely what the philosopher Friedrich Nietzsche captured by his 1895 maxim: “What does not kill me, strengthens me”. This is the fundamental principle, expressed in the perspective of the enemy, that governs the near-obligate comeback of terrorists after we destroy their facilities with bombs or the reemergence of regimes in more repressive form after failed coup attempts. But this principle also underlies the recurrence of cancer after we melt most of the tumor tissue away with irradiation, chemotherapy, or modern “smart drugs”.

These parallels between warfare and cancer research are not made-up, not off-the-cuff. They are real and profound — but here is the surprise to the political leaders and generals: The entire cancer research community has not learned these two lessons in the 40 years of unsuccessful attempt to cure cancer. This community includes the program directors at the National Cancer Institute (NCI), who steer the direction of cancer research, and the investigators in academic and industrial labs. The cancer research community is culturally and intellectually far behind that of political scientists and military leaders.

Now onto the specifics of these two unlearned lessons and what we can do. They are interlinked. The first lesson is a policy issue, dealing with the ability of recognizing and correcting a wrong direction from within a community and the second lesson is about the specific strategy for combating a resilient evil that is complex, reactive and adaptive.


The perpetual challenge of having to overcome myopic group-think and to evaluate unconventional strategies in politics and military is well-appreciated, and mechanisms to take on these challenges are well-established. But if the bad guys in the world are like a cancer, evasive and adaptive –why has the cancer research community not learned from the public discussions on policies and military strategies to articulate its failures and to consider the very possibility that there could be alternative strategies?

There are many reasons why the entire community of cancer research is inherently incapable of outside-the-box-thinking, of questioning established wisdom and embracing non-intuitive strategies. These epistemic deficits are further cemented by the self-fulfilling prophecy of group-think that is highly prevalent in a system driven by a peer-review system. The current biomedical research system (not only cancer research) fails to be innovative in countless ways — it is in a crisis (as reported elsewhere, for instance here, here and here and here).

But in this post I propose a rarely articulated, “deeper” reason –which perhaps is too philosophical to be actionable: The cancer research community, including governmental funding and the private to the academic research labs, is intrinsically not capable of scholarly thought and discussion of ideas that spring from the mind of OTHER PEOPLE than oneself. This hampers collective learning. For instance, unlike in other domains of science, as widely ranging as physics, psychology or sociology, in biomedical research there is very little history-awareness. And without knowledge of your own history you do not know yourself, let alone that you are a wrong track at a historical scale. Cancer researchers do not know where key concepts of cancer biology come from. There is little incentive to think about the fundamental essence of cancer the way other academic disciplines think about their object of study. This absence of epistemic awareness and the lack of intellectual sophistication to fathom foreign ideas is of course not because cancer researchers are inherently dumber than, say, social or political scientists or historians.

Instead, the anti-intellectual climate in biomedical sciences, which is particularly pronounced in cancer research, is quite unique in modern academia and can be traced back to the pressure of pragmatism: The focus on applications — but without the rigor of engineering sciences. Just take a look at the particular skills required to conduct biomedical research: meticulous attention to details and tedious repetitive and menial manual work in the labs, combined with the simplicity of underlying rationales. This is not different from manufacturing, just a bit more abstract. The intellectual requirements for understanding current concepts of cancer biology does not exceed that needed for, say, driving a car or using your smartphone. Very mechanistic, along obvious lines of causation, no depth.

The dominance of attention to operation and management rather than scholarly discourse about foreign concepts creates an atmosphere that does not attract the scholarly prone thinkers –those who have the mental facility to withstand the pressure of orthodoxy, step out of existing frames of thought, see the bigger picture and ask big questions… (What actually is a “big question”? Francis Crick, who co-discovered of the DNA structure, asked one: “what distinguishes the leaving from the dead?”. By contrast, a modern cancer researcher asks: “Which parts of the drug molecule should I modify to increase its affinity to its molecular target? — let’s try”). Cancer research is thus more akin to computer hacking than computer science. It can be methodological and obviously logical (“this gene is found to be overactive in cancer — so let’s try to target it with a drug”).

The pressure to produce short term returns within an existing, simplistic conceptual framework taken as THE ground truth, has led to a professional version of the infantile need for instant gratification. This unquestioned attitude prevents deep reflection and promotes quick reflexes. The reign of reflexes over reflection, of automatism of intuition over autonomy of intellect, has long spread throughout the entire land of academic cancer research, wiped out diversity and stifled critical thinking and curiosity. Luckily, the “soft” sciences, such as political, behavioral and social sciences that underlie public policies and military strategies, have preserved the ability of pure academic discourse and theorizing. This is why cancer researchers can learn from them.

All this is lucidly (sadly) manifest in the first reports of the task force and advisors of the presidential Cancer Moonshot program published this fall. Recommendations herein build on the same old ideas and lack any discussion about our stagnation in the war on cancer and about new ideas — like the public debates we see among policy makers and pundits. In the first report by the cancer moonshot taskforce, there is only a vague passionless commitment to promote “breakthroughs”. Most pages are filled with operational suggestions to implement existing ideas faster: to address logistical bottlenecks in technology deployment, data collection, data sharing, data analysis, clinical trials, etc. If there is no awareness even of the very possibility of the existence of a culture of intellectual pursuit, then there cannot be a dialectic discussion that is needed for collective self-criticism. With this endogenous inability for collective self-reflection, the cancer research community is trapped in a giant bubble of its own making and without being aware of it.

Thus, in stark contrast to the public discussions about sending more troops into war zones or extending embargoes, the cancer research community is not even aware that it is “staying the course” and “doubling down”: For, it goes without saying that we need to kill more and more cancer cells –with more precision and with more powerful drugs. All we need is to become better at spotting, classifying and killing cancer cells. Thus, cancer becomes an engineering problem that demands no more scientific inquiry, only working out the logistical problems. Really? But what if killing of cancer cells, however efficiently and precisely, is a bad strategy? (Don’t we know that drones and laser-guided bombs are not the sole solution in the war on terrorism?). This very possibility is as unfathomable to the cancer researcher as is the existence of stars to the cave fish.


This brings us to the second lesson that cancer researchers have not learned from political scientists and national security experts despite the omnipresent debate of the war on terrorism: The most plausible, most direct strategy to win against the bad guys, killing terrorists or imposing sanctions on dictators, can always backfire. It can make the enemy stronger. As said in the introduction, this is also the case with cancer therapy. Treatment procedures can directly promote the progression of a tumor to a more aggressive and treatment-resistant form by making some cancer cells more malignant instead of killing them. This is the Nietzsche principle applied to the tumor.

The Nietzsche Principle for cancer cells: Treatment with the intention to kill cells (blue) almost inevitably will cause some cells to instead convert into a stem-cell-like state (red). These are the seeds for recurrence of the tumor and treatment resistance. The therapy stress does not weaken, but strengthen the surviving cancer cells.

Unfortunately, only a small minority of cancer researchers has noticed this phenomenon (see e.g. here and here and here for original research examples). The vast majority of cancer researchers currently operate at the end of the ideological spectrum that is equivalent to extreme belligerence. They would make Dick Cheney or Donald Rumsfeld pale of envy. They think that because every new anti-cancer “smart drug” has, as it turns out, very limited efficacy (even celebrated breakthroughs typically extend life by typically less than a year), we need to double down: to quickly develop and deploy new drugs that are more lethal to cancer cells.

Of course the possibility that treatment eventually will backfire (except under rare favorable conditions) does not mean that one should always forgo the act of killing to save lives in the short term. Both in the war against terrorism and against cancer, lethal force is warranted to avert imminent danger — but it is always a doubled-edged sword to be wielded with due caution. Specifically, the tumor burden that can cause destruction of tissues and organs must be dealt with. But while most treatment can induce tumor regression, it is very difficult to prevent recurrence of invasive tumors even if the primary tumor melts away with the treatment.

Now let me explain the actual biology: why does cancer treatment backfire? Why is tumor recurrence not simply the regrow of left-over cancer cells missed by the beams or molecules of treatment? Why does treatment make the tumor cells that survive reemerge more aggressive, much as bombing campaigns emboldens and empowers the surviving dictators and terrorists? This is not an off-the-hand analogy. It has much to do with the inherent resilience of evolved complex adaptive systems, and very little with classical Darwinian evolution, the conventional explanation for tumor recurrence. In this traditional explanation, ingrained in the minds of oncologists, the genetically weaker cells are killed by therapy and the genetically stronger and resistant cells survive and pass on their genes, exactly as biological orthodoxy wants to have it. The stronger cells are passively selected for; individual cells do not change their functional properties in response to the drug.

In the case of cancer, new genome sequence analyses of tumors now suggest that there is little evidence for a major role of such Darwinian evolution among the cancer cells in a tumor. Instead, we propose that Nietzsche dominates over Darwin (or jump-starts it). Now let us look at the cellular mechanisms (for a technical review see here).


A tumor is not just a homogeneous mass of selfishly growing tissue. It consists of zillions of vastly diverse cells, each of which responds in vastly different ways to external biochemical signals, such as hormones and drugs. As a consequence, no treatment (except of course surgical removal of small localized tumors), be it chemo, irradiation or modern target-selective (“smart”) drugs, can reach every single cancer cell and kill them off all. This is well-known to cancer biologists and oncologists. But most think that the non-killed cells simply lay hidden and sometimes regrow after some while, driving the dreaded tumor recurrence.

BUT: It ain’t that simple! For one, the recurrent tumor is invariably more aggressive and more therapy-resistant –this is the chief reason for the ultimate failure of all these new fancy therapies that we read about in the popular media. Now, the key point is that this increase in aggressiveness in cells that have survived treatment can occur WITHOUT additional mutations in the surviving cells. And without mutations there is no evolution in the traditional Darwinian sense. This alternative mutation-independent process has been repeatedly proposed because recurrence can be so fast as to defy an evolutionary process but it has been ignored by almost all modern cancer researchers. They still believe that the only explanation is that tumor cells “evolve” more malignant traits in a process that follows the principles of Darwinian evolution. We now know that something much more sinister than the passive Darwinian selection of “fitter” mutant tumor cells is happening.

For, those cancer cells that were not killed by therapy, but whose neighboring cells may have succumbed to the drug, are not static bystanders that happen to be inherently more resistant (which would fit the traditional idea of a Darwinian evolution). These cells do not necessarily carry the genes that make them drug resistant. They just were luckier to have been in a state or position that allowed them to survive the therapeutic attack.

BUT: Surviving the therapeutic attack changed these cancer cells –much as YOU would be a different person after a near-death experience. The surviving cancer cells experienced a major cytotoxic stress. This is a massive perturbation of the cells’ internal systems (genome and cytoplasm) and automatically triggers the activation of cell survival programs that are wired in the deeper layers of the cells’ genomic control systems. Some of the genes involved in this response go back to those that had evolved to protect single-cell organisms. For instance, a cancer cell that has been stressed by a highly cyto-toxic chemotherapeutic agent will activate genes that make proteins that neutralize or eject foreign chemicals –very similar to those that protect microorganisms from toxins and heavy-metal. In fact, many genes activated in the stressed cancer (or normal) cells belong to a broader protective stress-response program. Interestingly, this cellular program overlaps with the one which affords cells some “stem-cell features”. This “stemness” program is important because stem cells are naturally more resilient to toxins, live longer (indefinitely?), are versatile and orchestrate embryonic development and tissue regeneration. But do not forget: we do not deal with normal benign stem cells but the rogue version in mutated cancer cells, hence the term “cancer stem cells”. A single cancer stem cell (by definition) can start an entire tumor. Thus, it is as if in response to the cytotoxic stress the tumor cell starts a pathological self-regeneration, perhaps even a new life cycle because “stemness” implicates a rejuvenation. This is eerily similar to the life cycle in primitive organisms, such as hydras (a coral-like polyp), where an environmental stress induces the production of stem cells, eggs and sperms, thereby starting a new sexual life cycle.

In view of this complexity of response in the cancer cells that survive the stressful event of being treated with anticancer drugs, it should by now be conceivable why Darwinian selection is not a suited explanatory framework for tumor recurrence: It can only explain the emergence of selectable traits, in this case drug resistance and faster growth — but not the coordination of a highly complex and coordinated process of rejuvenation and regeneration within the short time period of a few cell divisions.

BUT: Here is another twist to the mainstream focus on cancer cells as the basic units of cancer: the power of a regenerative response that tumors harness is not driven by the reaction of individual cancer cells to stress but is a well-orchestrated response by the cell population as a whole. Cancer is not a disease of the cancer cell but of the tissue, the proverbial “whole” that is “more than the sum of its parts”. Cells in our body talk to each other through a complex network of biochemical signals (“micro-hormones”). Now, in conjunction with partial injury of tumors, there is another eerie and forgotten fact: the injured, stressed, dying and dead cells –those that were no so lucky during treatment– send out “danger signals” to the surviving cells: They release intracellular molecules not found in normal tissues. These danger signals join the toxic molecules of the anticancer drug in stimulating the activation of the stem-like state in their surviving neighbors. In normal tissue, this is part of the regenerative program: when in the adult tissue (where cells do not divide anymore) a cell dies, leaving a hole in the tissue, neighboring cells rejuvenate and become stem-cells that can divide to fill the gap. Fittingly, the danger signals stimulate inflammation and wound-healing, and also suppresses the immune system that may interfere with regeneration. In fact, as has been known for over half a century, that in mice, injection of dead cells (killed by radiation therapy) into implanted tumors can drastically stimulate growth up to 100-fold!

We suspect that this response to injured or dead cells by neighboring cells are part of “tissue self-repair” programs in healthy organisms. But in the context of abnormal cells like cancer cells, these regeneration programs can go awry — as 19th century pathologists have recognized (when cancer research used to be carried out by careful observers and thinkers): A tumor is a wound that never heals.

In the war on terror it is well recognized that death in the own ranks sends reinvigorating signals to the entire society: Kill the kin of your enemies –and you ensure that your enemies will fight back with fiercer determination. (Only a fool would seriously suggest to kill the family of terrorists).

In summary, Nietzsche’s counter-intuitive concept that partial destruction does not make the enemy weaker but stronger, applies at two levels: (i) non-killed cancer cells emerge stronger because the cyto-toxic tress that they experience themselves during therapy triggers the stem cell program in the surviving cancer cells and (ii) the injured tumor as a whole emerges stronger because of the danger signals sent by the dying cells to their surviving neighbors trigger a inflammatory and regenerative response. Thus, tumor treatment does backfire and this is manifest as recurrence in more aggressive form. (Note that it never ever happens that a recurrent tumor is weaker and easier to treat but I know of no clinician who has ever expressed wonder about this.)

Classical Darwinian evolution, according to which cancer cells carrying mutations that make them more aggressive would be “selected for” (survival of the fittest), does have a role to play but evidence from genome sequencing of tumors suggest that in general it contributes only to a very limited extent to tumor progression, and mostly in punctual cases, such as in immune evasion against a specific immunotherapy or in altering the proteins structure of an oncogene targeted by a target-selective drug. Darwinian evolution among the cancer cells does not create the sophisticated (but pathological) regeneration program. In most cases it may act along the path paved by the powerful Nietzsche principle of reactive strengthening of cells in the stressed tumor tissues that immediately follows any cell killing and does not have to wait for the appropriate genetic alterations in the Darwinian lottery of mutations.

As with terrorists and dictators, cancer cells are “desperate” to survive. They are also resilient complex adaptive systems. “Doubling down” in combating such reactive systems is not likely to work. Mathematically, the Nietzsche principle implies that we do not have a zero-sum game: the damage inflicted do not add up linearly with the intensity or frequency of our attacks because the target is reactive. You cannot come back to finish off the killing.


Despite the complete failure to see that the Nietzsche principle is a major driver of tumor progression after cyto-toxic anti-cancer therapy, we have to acknowledge the good intentions. What current cancer researchers, clinicians and drug developers hope for is that by more specific and efficient at killing cancer cells, using more precision weapons, or clever combinations of them, or using self-powered and self-propagating cellular kill-vehicles borrowed from the immune system, one could exploit a narrow window of opportunity in which one may be able to cleanly kill off all the enemy cells quickly without risking making them stronger à la Nietzsche. This might in principle be possible and perhaps explains the rare cases of cure even of advanced cancer. But I suspect that a window of opportunity in which a zero-sum game exists is unrealistically narrow, if not fundamentally non-existent. So why not think outside the box of “more killing”?

That killing cancer cells can backfire because of Nietzsche is only the second lesson from the war om terrorism. But to ever reach this lesson let alone internalize it and convert into new treatment paradigm, cancer researchers must first learn the first lesson for which politics, public policy and warfare offers plenty opportunity: to acquire the general mental facility of collective introspection and correction. And here is the problem. To the overwhelming majority of cancer researchers, the space outside the sphere of their own thinking is unfathomable, void, non-existent. Cancer research has over the years ridded itself of any scholarly discourse and created a culture that stifles thinking beyond one’s own frame of thought. Of course, cancer researcher do think –but they think within the confines of IMPLEMENTING strategies that spring from one single unquestioned conventional wisdom.

The next challenge for cancer researchers is to first learn to think and to question conventional wisdom. To step out of one’s intellectual comfort zone, to learn that they are in a bubble and that there is something outside. This will open an entire new world of ideas, of which a specific one is the second lesson: The cellular conditions that cause treatment to backfire. Then we will be able to develop a new class of drugs that would specifically suppress that ancient cell stress-response and the run-away regenerative reaction of the tissue –that which makes the cancer stronger when we try to destroy them. Alternatively, we could also learn to “kill gently” to avoid triggering the Nietzsche principle. Perhaps this is what happens with immunotherapy: the natural elimination of unwanted, injured or dead cells without causing tissue damage as it happens during development and wound healing. But current immunotherapy, the most promising of all therapy strategies, also focuses on maximizing killing and “immune evasion” by the tumor is a major challenge. Even immunotherapy, hailed as a breakthrough, resides in the prison of the same culture of cancer research.

Thus, to defeat cancer we first need to defeat the establishment of cancer research.

Institute for Systems Biology

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store