Respectful Insolence

ResearchBlogging.orgWant to know what will start my teeth grinding when I read it in a newspaper? That’s easy. It’s headlines like this one, which appeared two days ago in The Telegraph:

Scientists two years from developing ‘potential cure’ for breast cancer

The subtitle was even worse:

British scientists could be just two years away from developing a drug that may be a “potential cure” for breast cancer, it has been claimed.

Hear that grating? It’s the sound of my teeth grinding together. The reason is simple. It’s just plain silly to make claims like this about a basic science paper given that, as I have discussed before, it often takes decades for basic science observations to wend their way through that long strange trip to becoming actual therapies used by clinicians. The life cycle of translational research is long, and efforts to speed it up have only met with mixed success:

Researchers have found that the cancer manipulates molecules, called microRNAs, to allow malignant cells to spread throughout the rest of the body.

The discovery means they can now work on developing a drug to stop this process.
Dr Justin Stebbing, senior lecturer and consultant medical oncologist at London’s Imperial College, is one of the experts behind the breakthrough.

“There are no available drugs as yet but they should be available within a couple of years. This is a potential cure for breast cancer,” he told the Daily Express.

“This is a step on the way to it and it helps us understand the way breast cancer cells grow and divide and if we understand this then we understand how it stop it.”


With all due respect to Dr. Stebbing, no and yes. No, this is highly unlikely to be a “cure” for breast cancer, any more than immunotherapy was in the 1980s, Herceptin was in the 1990s, antiangiogenic therapy was in the 2000s, or any of a number of touted “cures” that have cropped up over the years. It does not become a cancer scientist to make such statements. On the other hand, yes, this is interesting basic research that could well lead to new therapies for breast cancer in several years. But it’s not a cure; it’s not likely to lead to a cure quickly; and it really irritates the crap out of me to hear Dr. Stebbing claim that something like this is likely to lead to a cure that fast. Yet this claim has been reported under headlines like the one above and these:

Would it were all true!

So what is the study whose promise is being so hugely exaggerated? It appeared in PNAS1 and represents a report of the work of Dr. Stebbing’s group in concert with Gregory J. Hannon’s group at Cold Spring Harbor and involves a kind of regulatory molecule known as microRNAs. It helps that Dr. Hannon is one of the founding fathers of RNA interference and gene silencing, although he didn’t win the Nobel Prize for the discovery, as Professor Andrew Z. Fire at Stanford University, California, USA, and Professor Craig C. Mello at the University of Massachusetts Medical School did in 2006. This represented the shortest period of time from initial publication of a discovery and the winning of a Nobel Prize for it. RNA interferences (RNAi) was only discovered in 1998, and it’s the first time I can remember of a Nobel Prize being awarded for something that was discovered since I’ve been in science. Even better, it was awarded for something that I’ve become interested in, because over the last few years I’ve become very interested in RNAi, and, in particular, the short RNA sequences that mediate it.

In particular, I’m interested in one specific RNAi mechanism, that mediated by specific short sequences of RNA that are not transcribed into protein known as microRNAs. That’s what this paper is all about. MicroRNAs in general result from the processing of larger precursor RNAs that are cleaved to 21-23 nucleotide stretches that have a complementary sequence to specific messenger RNAs (mRNAs) responsible for coding for protein. What happens is that a microRNA binds to the complementary sequence in a mRNA and then either induces the degradation of that RNA or inhibits the translation of that RNA into protein. Either way, the end result is the same: The level of the protein that the mRNA targeted by the microRNA codes for decreases. In other words, microRNAs shut down gene expression, and they do it very specifically. Indeed, over the last few years, it’s become more and more appreciated that microRNAs are a very important post-transcriptional (after the mRNA is made from DNA) regulatory mechanism for controlling gene expression, or how much protein a given gene makes. Because microRNAs are so specific, it is not difficult to imagine trying to target genes using them to shut down a gene quite specifically, and, indeed, that’s exactly one area that scientists are studying. In addition, it turns out that microRNAs represent a very important regulatory mechanism controlling key molecules and that alterations in what microRNAs are or aren’t made can have a profound effect on the transformation of normal cells into malignant cells. They target multiple important cancer-related genes, and in some cases they are more accurate a predictor of tumor type than the genes they regulate.

What Dr. Stebbing looked at was the role of microRNAs in regulating the expression of the estrogen receptor, which is incredibly important in breast cancer. Basically, the estrogen receptor (ER) is a protein in the cell that estrogen binds to. After estrogen binds to it, the estrogen receptor moves into the nucleus and turns on various genes that cause the cell to grow, among other things. The reason ER is so important in breast cancer is because over 2/3 of breast cancers are what we call estrogen receptor-positive (ER+), meaning that they make the estrogen receptor and are therefore stimulated to grow by estrogen. That’s why the mainstay of breast cancer therapy, at least for ER(+) cancers, is antiestrogen therapy such as Tamoxifen or aromatase inhibitors. Unfortunately, only approximately 70% of ER(+) cancers respond to antiestrogen therapy. Worse, even those that do respond frequently become resistant to antiestrogen drugs. Antiestrogen resistance is a very serious problem in breast cancer; when breast cancers become resistant to Tamoxifen usually all that’s left is chemotherapy.

Stebbing observed that the activation of the ER results in the production of specific microRNAs that in turn target the mRNA for ER. Thus, the activation of the ER results in a feedback mechanism that attenuates and shuts down the cell’s ability to respond to estrogen, thus terminating the response before it can get out of control. In breast cancer, under estrogenic stimulus, an oncogene known as c-myc is activated, and c-myc in turn activates specific microRNAs. The interesting result of this study, at least as described by the authors, is that in ER(+) cells, for some reason the normal precursors of these microRNAs that target the ER are not properly processed into active microRNAs. In other words, for some reason the feedback loop that shuts down ER signaling doesn’t get turned on as it normally is, leading to continued stimulation.

These are all interesting results, but there are a number of problems and challenges that make any proclamation of a “cure” within two years to be premature at best and irresponsible at worst. First of all, the in vitro work was all done in a single cell line, MCF-7 cells, a breast cancer cell line isolated from cells in a pleural effusion of a woman who died of breast cancer back in the 1970s. Every experiment looks at either MCF-7 cells or MCF-7 cells that are genetically modified to do various things. There’s no way of knowing if Stebbing’s result is generalizable to most, many, or even a few ER(+) breast cancers in humans. In addition, there was no direct comparison of the levels of the two precursor microRNAs that correlated with ER levels between normal breast tissue and breast cancer, an important omission. True, the investigators did measure levels of the two microRNA precursors in a small number of human breast cancer specimens and found that they correlated with ER mRNA levels, but that’s more correlative and supportive, rather than particularly convincing. They also noted in many more samples that one of these same two microRNA precursors is expressed at a higher level in ER(-) breast cancer cells than in ER(+) cells, but, even though the result is statistically significant, I wonder how biologically significant it is, given that the difference is small and there is considerable overlap. I also wonder why the used so few samples to look at whether these microRNA precursors correlate with ER levels in ER(+) breast cancer when they clearly had many more ER(+) breast tumors that they could have looked at. Finally, in all of the cell culture, there were in essence no functional data presented, in other words, data that show that manipulating these microRNAs have biological effects at the cellular level that would indicated that, for example, increasing their level would shut down breast cancer cell proliferation or metastasis, and there was no animal data that suggested that the microRNAs described were functional in tumors.

As I said, this was a strong preliminary study, but it lacked functional data and animal studies.

Worse, the investigators go far beyond what their study shows when they speak to the press:

Scientists have previously focused on how to prevent tumours from forming, but the new research has found key molecules called microRNAs which the cancer manipulates to spread around the body.

Breast cancer cells ‘switch off’ these molecules, allowing the cancer to spread unchecked to other parts of the body.

This spread is responsible for 90 per cent of deaths from breast cancer and the team which made the breakthrough is working on a drug to stop this fatal process.

There was nothing–I repeat, nothing–in Stebbing’s study that looked at whether these two precursor microRNAs and the mciroRNAs that derive from them affect metastasis. The above statement is ridiculously overblown and completely unjustified based on the content of the paper. I hate it when scientists feed the press conclusions based on the data from their papers that reviewers would never have let them get away with in the actual manuscript. (Here’s a hint: “complementary and alternative medicine” practitioners do this all the time.) Stebbing would l do well to learn from the example of one of my scientific heros, Judah Folkman. Back in 1998, when he produced evidence far more compelling than anything in Dr. Stebbing’s PNAS article that he had found a highly effective treatment for cancer (after all, he had animal data; Stebbins does not), Dr. Folkman did not say things like, “This will lead to a potential cure for breast cancer in two years.” Dr. Folkman appreciated how difficult it is to translate basic science into actual clinical therapies. Stebbins would do well to remember a famous quote from Dr. Folkman, variants of which I personally heard him say in two different talks and which was described in this excerpt from a People article on him in 1998

Now, Folkman, 65, a researcher at Children’s Hospital in Boston, has offered hope to the world. Though he hastily counseled caution as news spread that his pioneering work with lab animals could lead to a breakthrough in curing cancer–”If you have cancer and you’re a mouse, we can take good care of you,” he said–the excitement of the moment was palpable.

Dr. Folkman knew that scientists had seemingly cured various cancers in mice many times before but that those results were generally not as dramatic in humans–if they even translated into human results at all. Consistent with Dr. Folkman’s caution, eleven years later antiangiogenic therapy has taken its place in our armamentarium of cancer therapies, but it is most definitely not a cure.

The bottom line is that Stebbing’s work is good science that was badly reported. It’s preliminary and basic at its core, which I like. It’s also nice, hardcore molecular biology and provides some tantalizing observations that might suggest a molecular pathway that could well be targeted to treat breast cancer. What it does not do is to provide a pathway to target that is likely to result in a cure for breast cancer in two years. I really wish it did, but it doesn’t. Maybe in a decade or so, this work might result in a usable drug targeting this pathway, somewhat sooner if everything goes very well indeed. Dr. Stebbins would do well to contemplate the example of Judah Folkman before talking to the press again. The hyperbole used to describe this microRNA research soured my enthusiasm for Stebbing’s work. Indeed, I smell a pitch for money for a startup company to develop a therapy based on this work.

REFERENCES:

1. Castellano, L., Giamas, G., Jacob, J., Coombes, R., Lucchesi, W., Thiruchelvam, P., Barton, G., Jiao, L., Wait, R., Waxman, J., Hannon, G., & Stebbing, J. (2009). The estrogen receptor- -induced microRNA signature regulates itself and its transcriptional response Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0906947106

Comments

  1. #1 wfjag
    August 27, 2009

    “Worse, the investigators go far beyond what their study shows when they speak to the press:”

    What part of “We need to secure additional grants to continue our research, so we’ll publicly hype our work worse than cheap streetwalkers so the public will call their elected representatives to support funding us” do you fail to understand?

  2. #2 Calli Arcale
    August 27, 2009

    And in other news, scientists are just two years away from inventing a teleporter/cloaking device/ray gun!

  3. #3 GeekGoddess
    August 27, 2009

    I was promised flying cars by now. They’re 9 years late.

  4. #4 Todd W.
    August 27, 2009

    @GeekGoddess

    I was promised flying cars by now. They’re 9 years late.

    Yeah. And they’re supposed to fold up into a light-weight briefcase at the push of a button.

  5. #5 Paul Browne
    August 27, 2009

    “British scientists could be just two years away from developing a drug that may be a “potential cure” for breast cancer, it has been claimed”

    It’s a pity that what people will remember won’t be “could be”, “may be” or even the unnecessary “Potential” but rather the “two years”. It’s just more ammunition for cranks.

    Given the very, very early stage this research is at two decades might be a more realistic time frame.

  6. #6 Catherina
    August 27, 2009

    my favorite is “Zebra fish could cure all illness” (see link under my name).

  7. #7 Pablo
    August 27, 2009

    I can tell you from experience that even if you do state very clearly that this work is very preliminary and the benefits are highly speculative, with a lot more work to do to develop it into a functional product (by others, even) people are still going to be asking when it will be available.

    We did a press release once. Never again. The public doesn’t want to know that you learned something really, really cool. They want to know how soon it can be used to cure them or they can make money off it.

  8. #8 ABradford
    August 27, 2009

    Knowing Fire and Mello’s work and a lot of the research and insights that have come out of it, I think microRNA does have a lot of potential, not just in cancer therapy. It also has a lot of potential pitfalls. Saying that the tools of microRNA treatments will start to become available within two years is a much more reasonable statement than that these tools are cures themselves.

  9. #9 Pablo
    August 27, 2009

    It’s curious, I just learned about “silencing” RNA (siRNA) the other day at a student’s oral exam. I thought it was a pretty amazing concept.

  10. #10 Mike Stanton
    August 27, 2009

    The UK papers today are full of stories about preventing mitochondrial diseases by genetically engineering ova with donor mDNA prior to fertilization. They are slightly less gung ho than the Telegraph story but fail to report any of the possible pitfalls that still have to be overcome.

  11. #11 Stagyar zil Doggo
    August 27, 2009

    Off topic, but very depressing news from Bangalore:
    Hospital turns to homeopathy over H1N1

  12. #12 David
    August 27, 2009

    Basic scientists often lack an understanding of what is required to go from identifying a mechanism to producing a working drug treatment in humans. At present, nobody knows how to give siRNAs to whole animals in a way that enables targeting the appropriate intracellular reactions – the RNA gets degraded in the bloodstream, or doesn’t get into the cell, or goes to the wrong organelle. Or its action in vivo is less specific than in the in vitro model, and it turns out to be highly toxic. Even if you’ve worked out a mechanism and even if you can deliver therapy, this is an area where calibrating the right dose and getting that dose in at the right dosing intervals will be a challenge. On top of that, you’d have to show safety in animals before going into humans. Pharmaceutical development takes an immense amount of time. And the sad fact is, almost every new experimental treatment fails, even when the basic science and animal studies look promising.

  13. #13 Jennifer B. Phillips
    August 27, 2009

    LOL@Catherina#6. That headline must be framed and prominently displayed around the University of Edinburgh fish labs :)

  14. #14 qetzal
    August 27, 2009

    The comparison to Folkman is indeed a great cautionary tale. Folkman’s group got stunning animal results with endostatin. They could completely inhibit growth of a wide variety of cancers in mice. All the pundits thought endostatin would turn fatal cancers into manageable chronic conditions.

    Endostatin was then licensed to EntreMed, and it was a complete flop in the clinic.

    In the Daily Mail piece, Stebbing is quoted as saying:

    We can use these microRNAs as a new treatment and make them do what current drugs don’t do.

    Someday, yeah maybe. But people like Stebbing who think it will be easy to turn miRNAs into drugs were obviously not paying attention during the great antisense boom & bust cycle of the 80s and 90s.

  15. #15 Sigmund
    August 27, 2009

    I work with microRNAs and cancer myself. In particular I am interested in altering their function as an adjunct to current treatments. The major problem posed by microRNAs in cancer is that the important ones are either upregulated to enormous copy numbers (up to 50,000 copies per cell – compared to 1 to 10 copies of a typical mRNA) or are downregulated from similar levels to very low levels. In practical terms this creates an enormous problem if you want to either replace or inhibit the malignancy associated microRNAs. There simply aren’t any current delivery systems that will allow for these type of copy number replacement or inhibition – particularly in notoriously hard to treat cancer tissues.
    That’s not to say its hopeless but the initial promise of microRNAs has been replaced by a more hard-nosed realization that there is still a lot to do.
    By the way Orac, I thought you were a little unfair on their possible motives. To me it simply looked like the standard sort of overblown claims made to push for more grant money.

  16. #16 Kemist
    August 27, 2009

    Dr. Folkman knew that scientists had seemingly cured various cancers in mice many times before but that those results were generally not as dramatic in humans–if they even translated into human results at all.

    One of the many reasons for that is IMHO, the type of model most labs use, i.e. tumor grafts.

    There are very good reasons to use it : it’s more homogenous, it takes less time than induced tumors.

    The problem with it is that it totally overestimates therapeutic responses.

    A grafted tumor has key characteristics which are different from an in-situ growing tumor. One, it’s foreign tissue for the host. This has led to many erroneous conclusions, particularly as to the efficacy of the immune system to destroy cancer cells.

    Two, many cells (typically millions) are injected at a time, generally in a nutrient gel, to insure that the graft will take. That does not reflect how a genuine tumor forms. This instantly appeared tumor is much more dependant upon neovasculariation than an in-situ growing tumor, which makes anti-angiogenics appear quite effcient indeed.

    These characteristics make a grafted tumor generally more fragile than an in-situ growing ones, and thus more likely to respond to a single treatment.

    And… that’s not even considering the fact that rats and mice can sometime react very differently to a drug than humans. Tamoxifen, for example, kills rats but not humans due to a difference in metabolism.

    A typical all-new drug takes five to ten years to develop from finding the target, that is, if your target actually end up as legitimate (and you can’t know that before having a workable drug candidate, which already demands a hefty investment) and the many dead ended development programs gathering dust in pharma companies basements are an eloquent proof that it’s not always the case.

  17. #17 Kriegger
    August 27, 2009

    Most people who have been diagnosed with chronic depression and anxiety have probably been prescribed modern anti-depression and anxiety medications such as Prozac, Celexa, Zoloft, Paxil and other SSRI related medications which seem to be the favorite choice of every modern doctor of the 21st century. Is this findrxonlinein article.

  18. #18 Random Physicist
    August 27, 2009

    “two years away” is a phrase common amongst researchers.

    It starts in grad school, when one’s Ph.D. is “two years away” for about 5 years.

  19. #19 DLC
    August 27, 2009

    Thanks for the write-up, Orac.
    Typical science ‘reporting’. Sensationalized headline followed by mangled version of the facts with a couple “money quotes” thrown in. The thing is, the story would have been just as interesting had it been written correctly.
    Oh, okay, it would have been interesting to me . . .

  20. #20 Rogue Medic
    August 27, 2009

    Maybe he thought he was being interviewed by the Huffington Post. They’ll believe anything.

    He should have said 2 minutes, rather than 2 years. HuPoo readers do not have long attention spans. Just like Leonard in Memento.

  21. #21 tsuken
    August 27, 2009

    OT, but I was shocked – shocked I tell you – at Orac’s advertising ;-)

    http://twitpic.com/fj4cs

    Srsly though, it is rather sad that the moment your website mentions woo, automatic ad server thingummies deposit a large load of excrement upon ye.

  22. #22 TB
    August 28, 2009

    David #12 nails one of the issues. Basic scientists often violate the rules of evidence they apply to their research when putting together the press release. In my field, some of the most embarrassing exaggerations come out of UC San Diego’s press office. You’d think an institution of that caliber would get it right.

    The blame for exaggerated claims in press releases lies with the scientists, because journalists rarely rewrite the press release for which the scientist is responsible, as DLC #19 implies.

    Pablo, #7, please persevere in learning how to use press releases properly. Just like any other technique you use in research, you mess up badly the first time if you don’t know the limitations and the tricks do doing it accurately. One helpful idea: writing has to be tailored to your audience, to do that well you have to know the audience and the context in which they see your work.

  23. #23 Jud
    August 28, 2009

    Orac wrote: “…when breast cancers become resistant to Tamoxifen usually all that’s left is chemotherapy.”

    Great to learn from the post and comments by knowledgeable readers. Now, if someone can spare a moment for an unknowledgeable reader, I’m very curious to know the difference between Tamoxifen and chemotherapy.

    Thanks.

  24. #24 Kemist
    August 28, 2009

    That’s not to say its hopeless but the initial promise of microRNAs has been replaced by a more hard-nosed realization that there is still a lot to do.

    I think one way to go is to make more resistant chemical analogues to micro-RNAs, not a micro-RNA per se but a micro-RNA mimic. That way you do not necesserily need to have that many copies delivered, and you will have less stability/formulation problems. You will find a lot of unemployed organic chemists out there to do that for you.

    By the way Orac, I thought you were a little unfair on their possible motives. To me it simply looked like the standard sort of overblown claims made to push for more grant money.

    Anyone else is remembered of DCA, the fabulous non-toxic wonder drug – at least in rats ?

  25. #25 martin
    August 30, 2009

    @Mike Stanton
    “The UK papers today are full of stories about preventing mitochondrial diseases by genetically engineering ova with donor mDNA prior to fertilization. They are slightly less gung ho than the Telegraph story but fail to report any of the possible pitfalls that still have to be overcome.”

    These guys are not exaggerating that much – the idea is quite old and what they present in their Nature paper is only the latest twist of the story, which importantly allows them to transfer nuclear DNA from the donor without any detectable mitochondrial DNA into enucleated host oocyte. This happened with nuclear transfer techniques, where significant contaminations with donor mtDNA were detected, which made them unsuitable for disorders caused by mutations in mtDNA.
    This is very different kind of a story than the one discussed by Orac and I believe that we may see the technique tried in humans within a relatively short time, provided they overcome regulatory hurdles.

  26. #26 Yaron
    August 30, 2009

    About science reporting and cancer “cures”, I did see this one from today which seems very appropriate (though in this case the scientist does seem to hold some of the blame):
    http://www.smbc-comics.com/index.php?db=comics&id=1623#comic

  27. #28 Chris
    August 31, 2009

    Of course, they’ve got the obligatory topless woman examining/cupping her breast. I’m just glad they don’t feel obligated to include photos every single time for colon cancer or prostate cancer.

  28. #29 Louise
    August 31, 2009
  29. #30 dee
    September 3, 2009

    “Great to learn from the post and comments by knowledgeable readers. Now, if someone can spare a moment for an unknowledgeable reader, I’m very curious to know the difference between Tamoxifen and chemotherapy.”

    Tamoxifen blocks the oestrogen receptor (a hormone) so its known as an endocrine treatment.

    Chemotherapy are things that are toxic to cells eg alkylating agents that disrupt DNA replication, or thingsl like doclitaxel that interfere with cell division.

    hth

  30. #31 Alejandro Montenegro-Montero
    September 3, 2009

    Hi Orac,

    Fascinating post.
    I selected it as one of my “picks of the week” of posts aggregated at RB in molecular biology, over at my blog http://researchblogging.org/post/gotourl/id/134127

    Cheers,
    -A

The site is undergoing maintenance presently. Commenting has been disabled. Please check back later!