[ed note: Michael Jorrin is a longtime medical writer (not a doctor) who covers health and medical topics and trends for our readers, generally without an investment thesis. His words are his own, enjoy!]
In the Doc Gumshoe blog about cancer that appeared in June of 2013 (you can check it out here), I commented on the “bad rap” that cancer care gets, which I thought was due to a number of factors, not least that, although all kinds of eminent figures including several Presidents of the USA have declared “war on cancer,” cancer continues to kill upwards of half a million of us every year, so clearly we aren’t winning that war. Or, at least, that’s the perception of a considerable proportion of the public.
Let me state it clearly and unequivocally: Doc Gumshoe is not going to announce that we are winning the war on cancer. The war on cancer is most likely unwinnable – in fact, even defining it as a war on a single disease entity is a big mistake and a set-up for disappointment and failure. However, if we look at cancer treatment as a continuous campaign on an immense number of fronts, we find that we’re doing surprisingly well on quite a lot of these fronts. More on that later, but here, for a start, are some recent statistics:
The National Cancer Institute (NCI) estimates for 2014 are 1,665,540 new cancer cases and 585,720 cancer deaths. This represents an increase of 26,630 new cases and 8,530 deaths over the 2012 estimates.
But the NCI also compiles what are known as SEER statistics (Surveillance, Epidemiology, and End Results). These are estimates, but they are generally regarded as highly accurate, based on a number of regional registries. The current SEER registry includes data on about 28% of the entire US population, which is a robust statistical sample. According to SEER data, cancer deaths per 100,000 population peaked in 1991 at 215.1, and have been declining annually since then. The SEER estimate for 2011 (most recent year available for SEER) was 168.7 deaths per 100,000 population, a decline of 46.4 deaths per 100,000 population. Based on the SEER data, this would imply an overall decline of about 26% in the cancer death rate. Yes, a decline!
The discrepancy between the NCI’s estimate that there will be an increase of 8,530 cancer deaths from 2012 to 2014 with the SEER statistics, which are not yet available past 2011 is largely explained by the fact that US population grew by about 5 million between 2011 and 2014, and at the rate of 168.7 deaths per 100,000 that would account for 8,435 cancer deaths – almost exactly the same as NCI’s estimated increase for 2014 over 2012.
But enough of this fiddling around with statistics. What seems solid and reliable about the SEER statistics is the 26% decline in the rate of cancer deaths (the rate – not the total number of deaths!) in the 20 year period from 1991 to 2011. This would imply that if the cancer mortality rate as of 1991 had continued unabated, another 100,000 or so people in the US would have died of cancer. Let’s chalk that up as definite progress, if not a major victory.
However, another factor that deserves consideration when we think about cancer mortality is the zero-sum aspect of the relationship between cancer and cardiovascular mortality. Death rates from that particular killer are down sharply. The American Heart Association calculates that if the CVD mortality rate had remained at its 1963 peak of 307.4 per 100,000 population, 621,000 additional CVD deaths would have taken place annually from about 1996 onward – that’s about 12 million people. But those folks who escape dying of heart disease don’t live forever – instead, they survive to succumb to something else, frequently the big C.
A similar relationship prevails globally. Countries with the highest life-expectancy also tend to have higher rates of cancer mortality, while those with lower life-expectancy have much lower rates of cancer mortality. The simple, brutal, explanation is that something else kills them first, such as infectious diseases. Most of Africa, India, and South-East Asia have much lower cancer death rates than Europe and North America. An exception to this link between lower life expectancy and lower cancer death rates is Russia, which has both lower life expectancy rates and much higher cancer mortality. This is likely due to extremely high smoking rates compounded with a non-functioning health-care system.
Uruguay stands out as a nation with high cancer mortality which is almost certainly related to smoking; cancer mortality in that nation of unrepentant smokers is the highest in the Western Hemisphere. Smoking is also a factor in high cancer mortality rates in several nations with excellent life-expectancy rates and also excellent health-care systems, such as Denmark and France.
According to the World Cancer Research Fund, smoking is a cause (a cause, not the cause) of about 85% of all cancers, as well (of course!) as the principal cause of lung cancer. Yes, yes, we know all about that. But new research now casts doubt on previous certainty about the causes of many cancers.
Is it getting cancer a matter of heredity, environment, or just bad luck?
I’m not giving anything away by jumping to the answer, which is, “Some of each.” But this recent, very careful and thorough research makes it look as though luck plays a much larger part than previously thought. Here’s a quick summary of the story, which was published in Science on January 2, 2015 (Science 2015;347:78-81)
A couple of Johns Hopkins researchers, Cristian Tomasetti and Bert Vogelstein, set out to calibrate the degree of association between stem cell divisions and cancer incidence in a number of body tissues. The thought behind this project is that every time a stem cell divides, there is a chance that it will mess up, such that the “offspring” cells are not exact copies of the “parent” cell. These inaccurate cell divisions are essentially random mutations, and the great majority of these mutations will be meaningless – the new cells will simply die or fail to replicate. In a few cases, the mutations will be beneficial, conferring some evolutionary advantage, and these mutations may become part of the species genome. But in some cases, the mutation will lead to cancerous cells, which continue to replicate and increase in number.
At the outset, Tomasetti and Vogelstein assumed that a certain proportion of cancers would turn out to be related to these random mutations, while others would be more closely related to inherited or behavioral factors. So what they did was measure the number of stem cell divisions in 31 types of tissue and correlate the frequency of stem cell divisions in each of those tissue types with the frequency of cancer in those tissues. Overall, they found that the correlation between the total number of stem cell divisions and total cancer risk was 0.804, which is considered to be a high degree of correlation (a correlation of 1.0 would mean complete correlation). In other words, the more stem cell divisions in a particular tissue, the more cancer.
As an example of the correlation of stem cell divisions and cancer incidence, they point to the frequency of colon and small intestine cancers. Colon tissue undergoes four times the number of stem cell divisions as small intestine tissue, and also is the site of a much higher rate of cancer. In contrast, in mice there are more stem cell divisions in the small intestine, and mice have a higher rate of small intestine cancers than colon cancers.
According to Tomasetti (a biologic mathematician), statistical theory says that the probability that causality can be imputed to a correlation is about equal to the square of the correlation. Since the correlation was 0.804, the imputed causality is about 64%, meaning that almost two-thirds of cancers are probably related to these random mutations following stem cell division. Please note, Doc Gumshoe is far from an expert in statistical theory, but in general it is thought that if a correlation is lower than 0.70, the imputed causality hovers around 50%, which means that causation is a big maybe.)
In terms of individual cancers, the researchers identified 22 as being primarily due to those random stem cell divisions, i.e., bad luck. The bad luck cancers included most brain cancers, lung cancers in non-smokers, esophageal cancers, many liver cancers other than those due to hepatitis C virus, cancers of the gall bladder, duodenum, and small intestine, ovarian and testicular germ cell cancers, and some leukemias. The cancers due to genetic and environmental/behavioral factors included basal cell carcinomas due to sun exposure, thyroid cancers due to radiation, lung cancers due to smoking, liver cancer due to HCV, some specific types of colorectal cancer due to genetic factors, and some others.
A shortcoming of the study was that breast and prostate cancer, which are respectively the most frequent cancers in women and men, were not included in the study because the researchers were unable to find sufficient data on the frequency of stem cell divisions in those tissues.
The study by the Johns Hopkins researchers has been attacked by a no less prestigious institution than the International Agency for Cancer Research (IARC), which asserts that the notion that “two thirds of cancers are due to bad luck” will deter the general population from modifying their behaviors and being vigilant about their real cancer risks. The reporting on the study in the media no doubt overemphasized the “bad luck” aspects, ignored the fact that breast and prostate cancers were not included, and failed to emphasize that in 9 of the 31 tissue types they studied, the investigators failed to find a correlation between stem cell division and cancer risk. But in my opinion, the study confirms what we knew already: that cancers arise in the body through a quite normal process of cell mutation, and are fortunately mostly kept in check through the activity of our immune systems. The mutated cancer cells that escape that process are the ones that cause trouble. And, of course, many other factors can contribute to the development of cancer, but pure chance certainly plays an important part.
Is this bad news, or might it also contain a snippet of good news?
For those of us who are convinced that clean living will prevent cancer (meaning no smoking, avoiding unhealthy foods, staying away from polluted environments, and boosting our own immune systems via whatever means we can) this is undoubtedly bad news. That applies also to those lucky folks with no close relatives who had cancer. I include myself in that cohort, and even though I am forced to acknowledge that I, too, could be struck by lightning, I continue to consider myself lucky.
"reveal" emails? If not,
just click here...
For the chronic worriers, it might actually be good news. It might be liberating to accept the reality that there are some things that are largely beyond our control.
But in yet another sense, it reinforces a Doc Gumshoe point, which is that a major weapon in the strategy to control cancer is early detection. If we can’t confidently prevent cancer by staying away from the known causes, we can at least try to detect it early enough to improve the chances of successful treatment. Here are some bits of data that support that point:
Going back to prostate cancer (which we’ve gone on and on about at great length), starting in the 1990s there was a huge increase in the diagnosis of prostate cancer, from somewhere around 100 cases per 100,000 population to well over 200 cases per 100,000. This was almost entirely due to the widespread use of the prostate-specific antigen (PSA) test, which was FDA approved in 1994 for routine screenings. (Note, the increase in prostate cancer diagnoses is not the same as an increase in actual prostate cancer incidence – it’s not that more men got prostate cancer, but that it was diagnosed in more men. But we can’t track the incidence of a disease without diagnosis, so from the standpoint of statistics, they’re the same thing.)
However, what’s revealing is that although more than twice as many cases of prostate cancer have been diagnosed, the mortality from prostate cancer has declined. According to the American Cancer Society, at least two million men in the US are now alive because their prostate cancer was identified and treated. Of all cancers, the discrepancy between number of cancers diagnosed and cancer deaths is greatest for prostate cancer – four times as many prostate cancers are diagnosed as prostate cancer fatalities.
The other form of cancer with a similarly wide discrepancy between diagnosis and mortality is breast cancer. As with prostate cancer, almost four times as many cases are diagnosed as there are breast cancer deaths. Another cancer with a relatively low fatality rate compared with the incidence/diagnosis rate is melanoma. At least some of the discrepancy between incidence and fatality rates for these cancers is likely due to the relative availability and simplicity of detection.
At the other end of the scale are the cancers from which result in the deaths of a high proportion of diagnosed patients: pancreatic, liver, and lung cancer, among them. A common feature of these cancers is that they are not easy to detect in the early stages. Thus, the emphasis in these cancers must be in modifying risk factors, when they are known.
So, if such a high proportion of cancers are due to the bad luck factor, when does modifying risk factors work?
One type of cancer where modifying risk factors does definitely work is liver cancer following infection with the hepatitis C virus. This downstream catastrophe following from a viral infection has led to the fierce and well-publicized competition between such outfits as AbbVie and Gilead, which you have certainly read about elsewhere. The two rivals are Gilead’s Harvoni, a combination of Gilead’s own Sovaldi (sofosbuvir) and ledipasvir, and AbbVie’s Viekira Pak, which combines a three-drug combination of ombitasvir, paritaprevir, and ritonavir, in a package with yet another antiviral, dasabuvir. Regardless of the deals made by the manufacturers with different insurers, both drugs will be highly expensive. But what the payers – and certainly the patients! – are looking at is that the one-shot cost of these drugs is small potatoes compared with the costs and consequences they are trying to avoid, including the costs of liver transplants.
Lung cancer gives us another example of the clear benefits of addressing risk factors, as well as the equally clear downside of ignoring risk factors. The critical risk factor in lung cancer, as you will have surmised, is smoking. The incidence of lung cancer in men reached a high of about 100 new cases per 100,000 men in the mid 1980s and since then has declined to about 60 cases per 100,000, as more and more men quit smoking. The decline in lung cancer rates in men will likely continue in future years.
The story with lung cancer in women is not so cheerful. The incidence rate continues to be lower, no doubt because fewer women than men smoke. But while the incidence rate in men was declining, it was going in the other direction in women, as more women picked up the smoking habit. In 1975, the incidence rate in women was about 25 new cases per 100,000. Around the time that the male lung cancer incidence rate was starting to decline in the 1980s, women’s incidence rate was still climbing, and fairly steeply, so that it had doubled by the year 2000, and it has only declined very slightly since then.
However, as I said earlier, avoiding risk factors only goes so far. Non-smokers get lung cancer, and people who never had hepatitis C get liver cancer. Time and chance happeneth to them all, sayeth Ecclesiastes.
A word about established treatment forms
Most of the improvement in cancer outcomes – but little of the noise in the media – is due to the established treatments, which have been slammed as “cut, poison, and burn,” or, in more polite lingo, surgery, chemotherapy, and radiation. All three of these modalities have been significantly refined and improved in recent years. A few words about each.
If the patient has a solid tumor which has not yet metastasized to other parts of the body, surgery is intuitively the most direct and effective approach: remove the tumor and the margins around the tumor, where some cancerous cells may be lurking. In some cases, this leads to complete remission without the need for further therapy of any kind; in some cases surgery is followed by some forms of chemotherapy. This approach to treatment has produced five-year survival rates of about 90% in breast cancer patients (for localized breast cancer the five-year survival rates are close to 100%, and for regional breast cancer about 80%), and higher than 95% in prostate cancer patients. Improvements in surgery for cancer have largely been the result of great advances in imaging techniques, which permit surgeons more accurately to locate and remove cancerous tumors. We should also not neglect the possibility that the surgeons themselves are becoming more skillful.
Chemotherapy is based on the somewhat primitive-sounding notion that cancer cells, being greedier than normal cells, will absorb more of a poisonous chemical and thus kill themselves. The adverse effects that come with chemotherapy are not pretty to look at (extensive hair loss) nor pleasant to experience, but in combination with surgery, chemotherapy can lead to long-term remission which can amount to the same thing as a cure. Chemotherapy is seldom employed by itself; it is generally coupled (or tripled) with surgery, radiation, or another form of drug therapy.
Radiation therapy has also benefitted enormously by improvements in imaging, permitting external beam radiation to pass through the target tumor precisely. Radiation therapy is now typically carried out in such a way that the beam of radiation passes through the patient’s body in different paths, so that only the target tumor is repeatedly exposed to radiation. This is accomplished by changing the position of the patient’s body with relation to the source of the radiation. Radiation is also delivered in the form of beads of radioactive material which can be delivered to the site of the cancer. SIR Spheres (selective internal radiation therapy microspheres) from SIRTeX Medical Ltd, and TheraSpheres, from BTG International, are tiny radioactive spheres, containing yttrium90, which are conveyed to the liver via small hepatic arteries. SIR Spheres and TheraSpheres are sometimes used for the treatment of certain forms of liver cancer. (I leave discussion of hepatocellular cancer, an exceedingly complex topic, to Dr KSS, but the delivery of radiation by means of these microscopic beads, rather than by needle implantation as in prostate cancer, strikes me as a promising mode of getting therapy right to the cancer site. This tactic takes advantage of the propensity of tumors to foster the growth of blood vessels that bring the nutrients the cancer cells need to grow and multiply.)
Newer approaches to cancer treatment that have entered the mainstream
I’m aware that citizens of Gumshoe nation have an ear to the ground for rumblings that may signal the Next Big Thing – i.e., the little biotech with a possible inside track on a multibillion dollar market. That’s much more Dr KSS’s forte than mine; what Doc Gumshoe is looking at here is strategies and mechanisms of action that have already produced some successful treatment modalities, as well as some strategies that are in the experimental phase. In my next piece I will try to identify some specific initiatives – drug names (if they even have names yet) and their developers, but I’m not competent in the least to give even a breath of a hint as to whether – or when! – the drugs will become hits in the market.
Inhibiting cancer cell replication
All cells, whether cancer cells or healthy normal cells, reproduce by splitting, or mitosis. A crucial difference is that normal cells have a built-in program that eventually kills them off. This is called apoptosis, and it serves a valuable function, which is to make way for new cells, so that most of the cell population in our bodies consists of young and healthy cells. Cancer cells are not subject to apoptosis and will go on reproducing endlessly, growing and spreading – that is, unless something prevents them from doing so. A number of drugs can inhibit this process, and these have demonstrated great clinical benefit in some cancers. Many of these drugs were initially derived from a substance found in the bark of the Pacific yew tree, called taxol. These include paclitaxel (Taxol, Bristol-Myers Squibb), docetaxel (Taxotere, Sanofi-Aventis), another formulation of paclitaxel (Abraxane, Celgene). Capecitabine (Xeloda, Genentech/Roche) also prevents cell replication, but by inhibiting DNA synthesis.
Another mechanism through which cell replication may be inhibited is targeting telomerase in cancer cells. (We mentioned telomeres in the discussion of the Mediterranean diet, which has been found to protect telomeres and thus extend the lifetime of healthy cells.) Telomerase is an enzyme that protects telomeres, which are the ends of genes, so targeting telomerase is a way of attacking cancer cell replication. An agent that has shown promise in that approach is imetelstat, from the biotech company Geron, which did much of the basic research into the role telomerase play in cancer. The 2009 Nobel Prize for Physiology or Medicine was awarded to two Geron collaborators, Elizabeth H. Blackburn and Carol W. Greider, along with Jack W. Szostak for the discovery of how chromosomes are protected by telomeres and the discovery of the enzyme telomerase.
Some drugs in the class called tyrosine kinase inhibitors also inhibit cancer cell replication. Tyrosine kinase is an enzyme which can play a number of roles, including stimulating the growth of new blood vessels (see below), but are also active in cell replication. In healthy cells, tyrosine kinase activates a kind of “on-off” switch that makes it possible for cells to reproduce, but in cancer cells, tyrosine kinase turns the switch on and keeps it on, so that cells just keep reproducing without cease. Imatinib (Gleevec, Novartis) is a cancer-cell specific tyrosine kinase inhibitor which is used to treat one form of chronic myelogenous leukemia (CML) and also gastrointestinal stromal tumors (GIST). CML median 5-year survival has nearly doubled since the approval of imatinib, from about 31% to about 59% currently, and median survival in patients with GIST is also currently about 5 years.
Inhibiting cancer cells from promoting vascularization
Cancer cells need nutrients to survive, and they get them by inducing the growth of blood vessels (angiogenesis). They release vascular endothelial growth factor, or VEGF, in amounts far greater than normal cells. A drug that targets angiogenesis through the release of VEGF is bevacizumab (Avastin, from Genentech/Roche). It binds to the VEGF molecule in such a way that VEGF cannot interact with its receptor on endothelial cells. Avastin is FDA-approved for a number of cancers, including colon, kidney, brain, and lung cancer; however, the FDA recently recommended that it no longer be used to treat breast cancer. Genentech/Roche are certainly not giving up on Avastin for breast cancer. Currently, 234 clinical trials with Avastin in patients with many forms of breast cancer are at various stages; most of these are in combination with other drugs.
In one form of breast cancer that is especially difficult to treat, tumor growth is fostered by another type of growth factor, termed HER2, which stands for human epidermal growth factor type 2. In turn, release of HER2 is triggered by tyrosine kinase in another of its cancer-related roles. Some drugs have been developed specifically to target tyrosine kinase in HER2 breast cancer, including lapatinib (Tykerb, GlaxoSmith Kline) and trastuzumab (Herceptin, Genentech/Roche).
Killing cancer cells with oncolytic viruses
The curious fact that some viruses preferentially attack cancer cells has been observed since early in the 20th century, and research into the possibility of putting this into therapeutic use began about 50 years ago, but only now are we getting close to approval of the first oncolytic virus-based treatment. BioVex, a small biotech outfit, developed a drug called OncoVex, based on the herpes simplex virus, for the treatment of melanomas. In 2011, Amgen made a deal with BioVex worth in total about $1.1 billion for development of the drug, talimogene laherparepvec, which Amgen renamed T-Vec. Clinical trials with T-Vec have been highly promising. The virus-based drug is injected directly into melanomas which are deemed to be not appropriate for treatment by surgery. One of the most impressive findings in the clinical trials with T-Vec is the relatively high durable response rate, meaning that melanomas treated with this agent tend not to recur, unlike melanomas treated by other means, which do tend to recur. T-Vec, along with some other oncolytic viruses, launches a two-pronged attack on cancer cells – not only does the virus lyse (destroy) cancer cells directly, but it prompts the patient’s immune system to attack the cancer cells as well.
Other oncolytic viruses currently under investigation include a recombinant adenovirus to target bladder, breast, ovarian, colon, and prostate cancers, from Cold Genesys, Inc, in California. Another California outfit, Genelux, is looking at attenuated vaccinia viruses, which have shown effectiveness in head and neck tumors and advanced solid tumors. (The vaccinia virus is used to vaccinate people against smallpox.) A Canadian firm, Oncolytics Biotech, has an agent, Reosyn, based on one of the reoviruses, which is now in Phase 3 trials against head and neck cancers and is being studied in lung, colorectal, and pancreatic cancers. And an Australian biotech, Viralytics Ltd, has a couple of agents in clinical trials, based on the coxsackie virus, against ovarian, lung, gastric, and pancreatic cancers. An advantage of some of these agents is that they don’t have to be injected into the tumor cell, but can be given intravenously.
All of these viruses have to be modified in some form so that they target cancer cells only and limit their potential for causing disease in the patient. A challenge for the developers of virus-based therapies will be to convince patients that it is safe to be treated by a pathogen which would otherwise make them sick.
* * * * * * *
I haven’t said anything about the many, many, many potential cancer drugs that are at various stages of development. We’ll save that for the next Doc Gumshoe blog. But, in preparation for that, I want to make a few points.
As in many medical treatment areas, a lot of the basic research in cancer is being done in academic centers, and a lot of the first-step work in drug development is being done by smaller biotechs. However, Big Pharma is involved up to the hilt. The three best-selling cancer drugs for 2013 were Rituxan, Avastin, and Herceptin, all from Genentech/Roche. Novartis is in there at number four with Gleevec, and Celgene, Lilly, Bristol Myers Squibb, and AbbVie are in the top ten. Amgen has six cancer drugs in Phase 3 clinical trials, and eight more cancer drug candidates in Phase 1. Many of these, and many others in development by Big Pharma, are drugs that aim to treat cancer through entirely different mechanisms than the ones described above, and many of them have been acquired by the big outfits after initial development was carried out by the little biotechs.
Many of those potential new drugs are based on genomics, i.e., analysis of the genetic characteristics of the specific cancer that is present in the patient, and also analysis of the patient’s genome: in other words, highly individualized treatment. In some cases, this can permit engineering the patient’s own immune response so that it specifically targets the cancer that has attacked the patient. The procedure is highly complex, time-consuming, and, needless to say, very expensive. Not all the new treatment modalities are based on genomics, but Big Data is big and attractive and more and more people know how to put it to use. Whether treatment options based on the individualized application of genomics can ever make a major difference in cancer morbidity and mortality is a question whose answer remains to be seen.
But what’s certain is that an immense amount of research is going on in the area of cancer treatment. Currently, there are 43,362 clinical trials in cancer treatment at various stages, from planning to completed. Some of these, I am confident, will yield cancer treatment options that will make a genuine difference.
I look forward to your comments. Many thanks in advance, Michael Jorrin (aka Doc Gumshoe)