The Laboratory of Peter P. Lee, MD

Laboratory focus: Studying the interplay between T cells and cancer

The immunotherapy of cancer holds promise of a new treatment modality which may be more specific and potentially less toxic than radiation and chemotherapy. To reach this elusive goal, we need a thorough understanding of the interactions between the immune system and cancer, and why the immune response fails to control cancer in the first place.

What role does the immune system play in cancer?

Myth 1 - Cancer cells are ignored by the immune system in cancer patients
It has been observed for decades that tumors are frequently infiltrated or surrounded by lymphocytes. However, the presence of tumor-infiltrating lymphocytes (TILs) did not correlate with clinical outcome. More recently, T cells have been expanded in vitro with IL-2 from approximately 80% of a variety of human tumors, including melanoma, RCC, breast cancer, ovarian cancer, and others [1]. T cells from approximately 30% of melanoma patients were shown to have specific, MHC-restricted cytolytic activity against autologous tumor cells [2], suggesting that these cells represent part of a specific immune response rather than a non-specific inflammatory response. Using peptide/MHC tetramers, we previously reported that circulating tumor-specific T cells can be found in around 50% of patients with metastatic melanoma [3]. On the humoral arm, a powerful method SEREX has been used to demonstrate that the majority of cancer patients also have antibodies reactive with tumor antigens [4].

Myth 2 - Immune responses to tumor cells are elicited only against unique 'tumor antigens'
Dogma in immunology states that T cells respond to 'foreign' antigens and spare 'self'. Thus, in cancer, it was believed that 'tumor antigens' must be novel antigens created by mutations or translocations. Ground-breaking work by the Rosenberg and Boon groups, and others, have shown that, in fact, most TILs are specific for self, non-mutated proteins (e.g. MART, gp100, and tyrosinase) [5, 6]. Hence, tumor immunity is autoimmunity [7].

Myth 3 - Tumor-specific T cells alone are sufficient for tumor regression
Clinical responses in cancer vaccine trials thus far have been disappointing. Much of this failure has been attributed to lack of successful induction of tumor-specific T cells. Of trials which incorporated immune monitoring, expansion of tumor-specific T cells has in fact not correlated with clinical response [8-10]. Even with adoptive cellular immunotherapy, in which ex vivo expanded tumor-specific T cells or TILs are infused into patients, clinical responders (mostly partial or minor) remain in the minority (15-35%) of patients [11-13]. Nonetheless, successful T cell responses do exist. In CML, we have found a strong correlation between the existence of a T cell population specific for a leukemia-associated antigen PR1 with favorable clinical outcomes in patients treated with IFN-g or ABMT [14]. Why tumor-specific T cells appear to provide a protective effect in certain settings but not in others remains unclear, and is the subject of active investigation in my laboratory and others.

Myth 4 - Tumor cells are passive targets for the immune response
There are a number of potential immunoregulatory mechanisms by which the body protects itself from autoimmune attack - being of 'self' origin, tumor cells may make use of any of these mechanisms. In fact, tumor cells have been shown to elaborate countermeasures against virtually every phase of the immune response. To evade immune recognition, tumor cells down-regulate target antigens and/or HLA expression [15, 16], as well as co-stimulatory and/or adhesion molecules expression [17]. Some tumors may actively suppress the immune response via the secretion of immunosuppressive cytokines, such as IL-10 and TGF-beta [18, 19], or via the expression of Fas ligand (FasL) to induce apoptosis of responding T cells [20, 21]. Global T cell dysfunction and alterations in T cell signal transduction molecules [22, 23] and in human cancer patients [24, 25]. While global immunosuppression often appears late in cancer progression, antigen-specific T cell anergy has been shown to be an early event in the course of tumor progression in mouse models [26]. We have found evidence for this phenomenon in human cancer. Tumor-specific, but not virus-specific, T cells were found to be anergic in vivo in patients with metastatic melanoma [3]. Lastly, the tumor microenvironment itself may be a formable barrier for an immune response - acidic pH [27], shed tumor gangliosides [28], and hypoxia/oxidative stress/glutathione depletion [29] have all been shown to cause cytotoxic effector cell dysfunction.

Challenges and Opportunities Ahead

A dynamic picture is emerging in which the immune system is far from quiescent in cancer, and tumor cells elaborate a host of countermeasures to evade immune destruction. In essence, these are two systems which co-evolve over time. Clinical outcome is ultimately a balance between the proliferation of tumor cells and the immune response which attempts to control these cells. Various treatments may perturb these two systems in different ways - in good clinical outcomes, the immune response ultimately dominates, while in poor outcomes, tumor cells dominate. It is becoming clear that the challenge in tumor immunotherapy lies not only in lack of tumor recognition, but also ineffectivity of the immune response that develops. While efforts to elicit tumor-specific T cells directly in patients via vaccination may overcome potential mechanisms of immune evasion by tumors cells, current immunotherapeutic strategies do not address their potential immunomodulatory mechanisms. Understanding the molecular mechanisms which underlie T cell dysfunction in cancer will be important to devise novel adjunct strategies to enhance or modulate T cell responses after tumor-specific T cells have been elicited. It will likely take a multi-faceted approach acting at several steps along the T cell activation and effector activity pathway to finally achieve success in cancer immunotherapy.

Key questions being pursued

  1. How does the immune system recognize a cancer cell from normal cells?
  2. What are the molecular mechanisms by which tumor cells modulate and suppress the host immune response?
  3. How to design novel treatment strategies which empower the immune response to co-evolve with cancer and ultimately win?

Powerful, state-of-the-art tools are used in the laboratory

  1. Peptide/MHC tetramers to identify and isolate tumor-specific T cells from cancer patients which develop endogenously, post-vaccination, or other immunomodulatory treatments.
  2. Extensive biological characterization of tumor-specific T cell response using 10-color FACS analysis, sorting and functional (cytotoxicity) assays, and gene expression profile analysis using DNA microarrays.
  3. Key is to correlate biological properties of anti-tumor immune responses with clinical outcome to understand what makes an immune response to cancer effective.

Mathematical modeling and computer simulations to understand the dynamics and kinetics of the T cell response in cancer
We are collaborating with faculty from mathematics to devise models to describe and simulate the interactions between immune and cancer cells. We are using systems of ordinary differential equations (ODE) and cellular automata (CA). Our early progress are revealing interesting non-linear dynamics which may shed new biological insights. We invite mathematicians interested in biology, as well as biologists with strong mathematical backgrounds, to join is in this exciting endeavor.

Peter P. Lee, MD: Publications