Cancer cell attack (from XVivo)Does the immune system control tumors?

The current understanding says “Yes”, but with reservations. As I’ve noted in previous posts (here and links therein, among others), there’s pretty solid evidence now that the immune system controls tumors in their early development.

Probably (we don’t know this for sure, but evidence points to it) there are many proto-tumors that begin to form, taking a few steps along the long route to full-blown cancer, but that are destroyed by the immune system long before they ever become detectable. Probably many more tumors form and take those early steps, and though they are not completely eliminated by the immune system they are controlled — the immune system prevents the proto-tumor from ever becoming more than a little cluster of cells, even though that little cluster of cells may persist for many years.

By the time a cancer is clinically detectable, though, does the immune system have any effect? Again referring to the current model, proto-tumors are able to advance to the detectable stage because they have avoided immune control. Therefore, the tumor we see are almost by definition uncontrollable by the immune system, right?

Immune control of clinical tumors?

T cells infiltrating tumor Actually, that’s not quite what the theory suggests. A tumor that’s reached the detectable level grows faster than the immune system shuts it down, true; but that doesn’t mean there’s no influence of the immune system. Yes, the tumor could be growing twice as fast as it should, with no influence of the immune system. But equally, the tumor could be growing 10 times too fast, with the immune system destroying 90% of that. The overall rate would look the same; but in the latter case, we only need to push the growth rate down, or crank up the immune response, by 11%, to drive the tumor into remission.

Is there any direct evidence that the immune system slows the progression even of outright cancer? Certainly there is, though most of the evidence I know of it a little circumstantial. One line of reasoning is that, if the immune system controls tumors, then we should see a correlation between immune responses to tumors, and their prognosis. In fact, there are quite a few papers that show that: For example, a paper in Clinical Cancer Research last month.1

This group actually looked at two parameters, that might or might not be connected, and their influence on prognosis of ovarian carcinoma. On the one hand, they looked at evidence for tumor immune evasion: How stringently was the tumor avoiding cytotoxic T lymphocyte recognition? On the other hand, they looked at infiltrating T cells in the tumor: How well could T cells recognize the tumor?

(To my mind, the latter is a much more important question, because we don’t know much about thresholds and cumulative effects of immune evasion — that is, we aren’t yet able to look at the recognition molecules as such, and declare that T cells will or will not recognize the tumor. Of course, this sort of study, that correlates phenotype and function, will be critical for answering that question.)

Immune evasion is bad for survival

Tumor and T cellsImpressively, there’s a strong link between good prognosis and phenotype. Tumors that seems to have good antigen presentation, have a better prognosis than those that have apparently blocked their antigen presentation pathways efficiently. (They were able to break it down further than that, to the specific types of molecules that may be important.) And these are not trivial differences; people with defective antigen presentation survived for 1 or 2 years, those with good antigen presentation averaged 4 or 5 years or longer.

Patients with all five markers positive in the tumor lived almost four times as long (median survival 5.67 years; P < 0.01) and were 4.74 times less likely to die from their disease

The other half of the study was almost equally impressive.

Patients with complete absence of tumor-infiltrating T cells were 2.04 times more likely to die from their disease (95% CI, 1.35-3.07) than those with one or more T cells (P < 0.01; median survival 1.67 years versus 3.79 years). … [ However,] Although peritumoral presence of CD3+/CD8+ T cells was a significant survival factor in the univariate analyses limited to patients with advanced-stage cancer, it did not emerge as a significant factor in multivariate analyses.

This sort of study can’t definitively answer the question of whether there’s any significant control of clinically detectable cancers. For example, since there’s evidence that chemotherapy success is linked to the immune response, perhaps the immune parameters here are actually measuring the efficacy of chemotherapy, and the immune response is ineffective on its own. Still, it’s certainly encouraging — it suggests that the immune system really is a potential partner in treatment of many tumors, and maybe gives a pointer to which tumors are more or less likely to respond to treatment.

  1. Han, L.Y., Fletcher, M.S., Urbauer, D.L., Mueller, P., Landen, C.N., Kamat, A.A., Lin, Y.G., Merritt, W.M., Spannuth, W.A., Deavers, M.T., De Geest, K., Gershenson, D.M., Lutgendorf, S.K., Ferrone, S., Sood, A.K. (2008). HLA Class I Antigen Processing Machinery Component Expression and Intratumoral T-Cell Infiltrate as Independent Prognostic Markers in Ovarian Carcinoma. Clinical Cancer Research, 14(11), 3372-3379. DOI: 10.1158/1078-0432.CCR-07-4433[]