Tags
Acute infection, Coley's toxins, Myeloid-derived Suppressor Cell (MDSC), Tumor-Infiltrating Lymphocytes (TIL), William Coley
Answer by Tirumalai Kamala:
Let’s first make explicit the attributes of the immune system at play here. Either the immune system does not respond at all to tumors (true tolerance) or it responds (not tolerance). However, are these the only options?
The word Tolerance inhabits a grey area within the immunology lexicon. Immunologists frequently use the word tolerance to denote absence of immune response when often a careful examination of their experimental methodology reveals that their actual intention was to denote absence of anticipated immune response, typically a response they routinely examine using their lab’s immunology toolkit. Some readouts are observational such as CD4 T cell responses, others are functional such as lysis by neutralizing antibody or cytotoxic CD8 T cells or change in size or number of experimentally induced tumor. Further, many immunologists also use tolerance to denote certain classes of immune responses such as those associated with tissue or organ transplant acceptance or those associated with mucosal sites or those associated with tumor growth. Again, these are not absence of immune response. Rather, they are presence of certain types of immune response.
Important as these distinctions are, let’s expand the conceptual landscape beyond these technicalities and make even more explicit the attributes of tumors and immune function at play. Does a tumor grow in spite of an effective immune response against it or because it successfully subverts such immune responses? Parsed in this manner, there appear two possibilities for tumor survival and growth. Either immune function tries to get rid of the tumor but fails, or the tumor co-opts immune function to survive and grow. Data support both possibilities, suggesting we could tweak immune function and harness it to rid the body of tumors.
I’ll explore two themes in this answer. One, since we typically assume immunity deals with infections, what is the association between infections and tumors? Do infections promote or inhibit tumors? Two, once a tumor establishes, what approaches does it take to co-opt or keep immunity at bay?
Do Infections Promote or Inhibit Tumors?
What is different about the immunological history of people who don’t develop tumors? Is it possible such individuals make effective anti-tumor immune responses and get rid of them? If so, could epidemiological studies reveal what is different about them? A trove of such studies suggest that a history of acute infections could endow us with better capacity to prevent or eradicate tumors, in contrast to chronic infections or inflammatory syndromes (1, 2, 3, 4, 5, 6). What types of infections confer anti-tumor protection? Infections increasingly rare in industrialized countries, such as pulmonary TB, pneumonia, Staphylococcus aureus, and other infections associated with fever over 38oC. These sound suspiciously similar to the factors that William Coley empirically associated with spontaneous tumor remissions (7, 8). Here are some of these epidemiological studies linking a history of acute infections with resistance to tumors.
1. Babies who acquire acute infections at day-care centers appear to have lower risk for Acute Lymphoblastic Leukemia (ALL) (9).
2. Inverse relationship between childhood history of measles, chickenpox, rubella, mumps, pertussis with risk for Chronic Lymphoblastic Leukemia (CLL) (10).
3. Dairy farmers are apparently less likely to develop lung, bladder, pancreatic or esophageal cancers (11, 12), with higher exposure to cattle conferring greater protection, and protection waning when the farmers switch to other occupations. Some agriculture and cotton textile workers seem less likely to develop lung cancers (13). Here the protective exposure is different in kind to those in points 1 and 2 since it needs to be continual to confer anti-tumor protection. These studies suggest that certain occupational microbe(s)/microbial product(s) exposure could continually modulate immune function to confer anti-tumor protection but that they don’t induce a strong and stable immunological memory of the type that rids tumors.
4. Some vaccines may mimic the acute infection effect on immune function, and either train it appropriately and/or generate anti-tumor cross-reactivity. BCG, when given early in life and found to be protective against TB, also seems to protect from leukemia (14). So do the small pox Vaccinia and Yellow Fever vaccines (15, 16, 17, 18). I don’t think it is coincidence that all three are live organism vaccines. Robust and stable immunological memory appears to depend on constant exposure to live organisms.
While there are caveats and disputes regarding data interpretations associated with many of these epidemiological linkage studies, they embody the resurgence of interest in Coley‘s ideas, and perfectly complement the current renaissance in tumor immunology. How far we have come in 3 short decades! Over the course of the 20th century, with radiotherapy and chemotherapy touted as anti-tumor breakthroughs, clinical investigations of biological remedies requiring stricter regulatory compliance, and poor patenting prospects of Coley’s toxins, his revolutionary insight of harnessing immunity to rid the body of tumors faded to the extent that, as recently as the 1980s, studying tumor immunity was considered, “a seedy intellectual neighborhood of fantasy and wishful thinking, a landscape littered with the hulks of abandoned hypotheses and charred reputations” (8). We are also left with the key question. Are these associations causal or casual? In contrast to Coley‘s time, today, rather than directly plunging into human experimentation, we need to await animal model data for definitive proof of causality linkage between acute infections and resistance to tumors.
What does a Tumor do to co-opt or keep Immunity at bay?
What about tumors themselves? Do they reveal presence of immune responses? Solid tumors harbor many immune cells. These include lymphocytes (TIL; Tumor-infiltrating lymphocytes) and Myeloid-derived Suppressor Cell (MDSC). These days, there are too many mouse model or in vitro human studies exploring these and other cells in ways that seem to me either too archaic, too boring or too esoteric. Once in a while, a study pops up exploring such an interesting idea that I remember it years later. Such a study was a mouse model of prostate cancer published in 2005 (19), which was among the first to show that cancer cells can impose such strong selection pressure on the surrounding tissue cells in their micro-environment that the latter end up hopelessly compromised in their function, helplessly in thrall to the tumor, a Trilby (novel) to the tumor’s Svengali. In this study, the tumor appeared to promote accumulation of specific http://en.wikipedia.org/wiki/Fib… that had undergone changes in their p53 gene and proliferated locally to create a micro-environment that promoted the tumor’s growth. p53 is a tumor-suppressor gene that acts as a break on cell proliferation. Its loss indicates a runaway cell, one now endowed with the capacity for unchecked proliferation. Data from such studies beg the question what is a tumor? Can a tumor be so neatly, surgically, strictly delineated as we continue to believe? Does the distinction between tumor and normal tissue become more nebulous as the former grows and spreads? If a tumor co-opts normal cells to this extent, is it even conceivable that such co-opted cells go back to normalcy once the tumor is removed, whether surgically or by therapy or both?
If the tumor is a plant, and the tumor micro-environment the soil (20, 21, 22), isn’t this scenario akin to Monoculture, where, in the wake of the havoc wreaked by the tumor plant, the tissue soil is depleted, denuded and no longer normal itself? If such is indeed the case, is it any wonder that strong anti-tumor immunity alone isn’t enough? With the tumor and co-opted tissue micro-environment cooperating to impose such strong selection pressure on anti-tumor immune cells, the latter also end up compromised, becoming TILs that cannot lyse tumor cells or MDSCs that inhibit TILs from trying to lyse tumor cells. Presence of immune cells in tumors is evidence of strong anti-tumor immunity alright, only it is not effective anti-tumor immunity. Since many Epigenetics changes are stable, such epigenetically changed tumor micro-environments may not only retain tumor-supporting propensity but also retain capability to subvert anti-tumor immunity . This suggests that such epigenetically modified tissue sites could continue to remain at increased risk of tumor recurrence even after a primary tumor is successfully removed.
Examined in this fashion, it seems clear that targeting tumor cells, tumor antigens and their blood supply alone won’t be enough to eliminate tumors. We must harness the surrounding tissue as well. And while we are at it, as part of the tumor eradication process, shouldn’t we consider approaches to re-train/re-educate the TILs, the MDSCs and other tumor-associated leukocytes? If the tumor induces in these immune cells stable epigenetic tumor-supporting changes, which it likely does, aren’t these cells likely to sustain tumors in the future as well, as and when such tumors arise, particularly in the same tissues as the original tumor? In other words, we also need to specifically target and neutralize the tumor’s efforts to co-opt the surrounding tissue micro-environment (23, 24) and anti-tumor immunity. We now better appreciate this imperative (25). To quote the authors of a recent review (26), “inhibition of chronic inflammation in the tumor micro-environment should be applied in conjunction with melanoma immunotherapies to increase their efficacy“. Such an approach will necessitate increasingly personalized approaches since tumor micro-environments tend to be heterogeneous, and vary from one person to another and from one tissue to another (27, 28).
Bibliography
1. http://www.biomedcentral.com/con…
2. http://www.dewaarheidovervaccina…
3. http://www.ncbi.nlm.nih.gov/pubm…
4. Krone, B., et al. “Impact of vaccinations and infectious diseases on the risk of melanoma—evaluation of an EORTC case–control study.” European Journal of Cancer 39.16 (2003): 2372-2378.
5. http://www.ncbi.nlm.nih.gov/pmc/…
6. http://www.ncbi.nlm.nih.gov/pmc/…
7. http://www.ncbi.nlm.nih.gov/pmc/…
8. http://www.ncbi.nlm.nih.gov/pmc/…
9. http://www.ncbi.nlm.nih.gov/pmc/…
10. http://onlinelibrary.wiley.com/d…
11. http://aje.oxfordjournals.org/co…
12. http://www.sjweh.fi/show_abstrac…
13. http://www.ncbi.nlm.nih.gov/pmc/…
14. Grange, J. M., and J. L. Stanford. “BCG vaccination and cancer.” Tubercle 71.1 (19titivity. 90): 61-64.
15. http://www.ncbi.nlm.nih.gov/pubm…
16. http://www.nature.com/jid/journa…
17. http://www.madskamper.dk/upl/web…
18. Mastrangelo, G., et al. “Does yellow fever 17D vaccine protect against melanoma?.” Vaccine 27.4 (2009): 588-591.
19. http://ac.els-cdn.com/S009286740…
20. Paget, Stephen. “The distribution of secondary growths in cancer of the breast.” The Lancet 133.3421 (1889): 571-573.
21. Fidler, Isaiah J. “The pathogenesis of cancer metastasis: the ‘seed and soil’ hypothesis revisited.” Nature Reviews Cancer 3.6 (2003): 453-458.
22. http://www.ncbi.nlm.nih.gov/pmc/…
23. http://www.hindawi.com/journals/…
24. http://www.ncbi.nlm.nih.gov/pubm…
25. http://www.ncbi.nlm.nih.gov/pmc/…
26. http://www.ncbi.nlm.nih.gov/pubm…
27. http://clincancerres.aacrjournal…
28. http://ac.els-cdn.com/S092544391…
What is the relationship between tumors and immune tolerance?
immunobiolog said:
there is a third possibility for tumor survival and growth, namely that immune response is normally blind to tumours (Mol Med Today (99) 5, 14–17
LikeLike
Tirumalai Kamala said:
Thanks for sharing that reference (http://www.sciencedirect.com/science/article/pii/S1357431098014075)
This idea simply re-words the notion of immune privilege, that some tissues (e.g. brain, eye, hamster cheek pouch) are so delicate, our immune responses could destroy them hence immunity is kept actively and deliberately away from them. Three things about that.
One, what happened to the idea of immune privilege? As we developed more sensitive tools to probe tissues and broader approaches to assess different classes of immunity, the idea of immune privilege could no longer be supported by data. Immunity wasn’t being kept away from these sites. Rather these tissues appear to support at default a type of immunity different from what’s usually assessed. For example, mouse model work by J. Wayne Streilein and Joan Goverman are useful guides for re-interpreting eye and brain immune privilege as types of immunity different from say IFN-gamma dominated.
Two, if immune response is ‘blind’ to tumors, how to explain that solid tumors typically have TILs (Tumor Infltrating Leukocytes) and MDSCs (Myeloid-Derived Suppressor Cells)? Rather than being blind, anti-tumor immune responses appear to be/made ineffective.
Three, let’s consider the evolutionary implication of immune privilege, i.e. our immune response being ‘blind’ to some part of our body. It implies that some tissue in our body could or should stay immunologically unprotected. To be successful over evolutionary time, an organism has to be able to live long enough to leave progeny behind. Seems like having a part of the body stay immunologically unprotected would be exactly the kind of weakness that would be exploited over evolutionary time by pathogenic forces, exploited to the detriment of species survival. This hasn’t happened to us so far.
Glad to get your comment. It gave me the opportunity to elaborate further.
LikeLike
immunobiolog said:
Disagree that it rewords immune privilege. There is no ‘choice’ made: if an antigen gets into a DC (expressing a co-stimulatory signal), there is a response to this. The control is that the response is transient and can be shut off by check-point inhibitors (PD1, CTLA-4…, hence MDSC): TILs reflect some (incidental/accidental) capture of Ag resulting in activated T-cells that are then shut off (IL2 reactivates?).
The blind concept is supported by the masses of literature that shows that ANY way of getting a tumour antigen into a DC gets a response (even RNA pulsing); hence the ability to respond is there but just not activated ((ie antigen capture key).
LikeLike
Tirumalai Kamala said:
‘There is no choice made’. By whom?
‘hence the ability to respond is there but just not activated ((ie antigen capture key)’ ‘hence MDSC’? Rather why MDSC in tumor if immune system is blind to it. Why is response transient? SC part of MDSC implies active agency on part of tumor to subvert immune response, be it PD1, IDO, etc.
‘Any way of getting a tumor antigen into a DC gets a response (even RNA pulsing)’? Precisely RNA pulsing! RNA in cytosol is not normal, especially inside the cytosol of a cardinal sentinel cell like DC. That’s why DC pulsing with RNA can drive an immune response.
Get a response. So what? To get a(n) (immune) response is one thing. Getting an immune response that prevails in eliminating a tumor is another thing entirely.
Why do you think tumor antigen capture is incidental/accidental?
LikeLike
immunobiolog said:
Choice by immune system, should I respond or not? Hence Ag in DC also expressing CD86/80 gives response, irrespective of nature of antigen.
SC part: Q for suppression is specificity. If one removes IL2, many (most?) T-cells stop responding: is this suppression or ‘anergy’ (or ‘quiescence’)
but detection of cytosolic RNA has no specificity for antigen: the specificity of the response suggests acquired not innate immune response
Agree response needs to clear tumour but this is less difficult than we think as the data on checkpoint block inhibitors shows.
Tumour antigen capture is (usually/normally) incidental/accidental as tumors grow as cells and DCs do not capture antigens from whole cells.
LikeLike
Tirumalai Kamala said:
Isn’t it more complicated than just CD86/80 expressing DC inducing response? For example, if responding T cell has higher levels of CTLA4, doesn’t matter how high the CD86/80 on DC or what Ag it presents, T cell response is turned off.
Also, don’t MDSC and TIL in tumor already tell us we are beyond the respond or not stage? By response, I don’t mean an antigen-specific T cell response. Normal tissues don’t have this extent of leukocyte infiltration. Question then is not whether there is anti-tumor response or not. Rather question is two-fold, one, how a given tumor subverts immune response against itself (either suppresses or renders ineffectual) and two, how to maneuver around such manipulation.
Solid tumors tend to have necrotic cores. Aren’t they a plentiful source of antigen?
Not sure that the current craze for monotherapy using mAb against individual checkpoint block inhibitors will fully pan out. Simultaneous multi-target approach will likely be more fool-proof.
LikeLike
immunobiolog said:
CTLA-4 (and others) are ‘activation antigens’ hence expression means T-cell has been activated.
agree with the 2-fold Q, though would argue that ‘subversion’ may just reflect normal mechanism to limit CTL effects (hence ‘activation antigens’ now ‘checkpoint blocks’
Necrotic cores? Need DC access and cell lysis for antigen capture. Thought cell death in solid tumors largely hypoxia which gives apoptotic cell death.
Interesting that checkpoint block inhibitor use works at all AND results in neo-antigen broadening. Agree multi-targeting the key (MolMedToday 2003 Dec;9(12):515-6)
LikeLike
Tirumalai Kamala said:
Understanding of cell death has moved beyond binary apoptotic, i.e. shorthand for non-immunogenic and necrotic, shorthand for immunogenic. For e.g. in necroptosis, necrotic death is non-immunogenic.
As well, regardless manner of death, physiological outcome in the environs also depends on how quickly and quietly dead cell debris is cleared away by the phagocytic reticuloendothelial system. With uncontrolled cell growth and associated dead cores a hallmark of solid tumors, stands to reason phagocytotic clearance pathways in a tumor’s environs are likely to be overwhelmed. So less likely tumor-associated dead and dying likely to be immunologically ignored.
LikeLike