Cassian Yee

The overarching goal of our research focus is to develop cellular immunotherapy into a treatment modality for patients with cancer and to foster this goal through translational, developmental, basic and applied basic research in immunology. In general, the research focus in our lab can be divided into clinical studies, translational/developmental studies and basic studies in T cell immunity.

Since 1995, our lab has been developing strategies to isolate and expand tumor‐associated antigen-specific T cells for treating patients with solid tumor malignancies and to identify the intrinsic and extrinsic factors influencing T cell survival, function and efficacy. To ensure that we would be able to more precisely evaluate the reasons for success or failure of a given approach, we chose to use T cell clones from the start. We postulated that a well‐defined population of T cell clones of uniform specificity, phenotype and magnitude would be instrumental in addressing immunologic questions and provide a highly functional anti‐tumor effector population for adoptive therapy.

To this end, we developed a strategic plan to systematically evaluate intrinsic (cell‐specific) and extrinsic (host‐specific) factors that could be modulated to enhance T cell persistence and antitumor efficacy (Hematol Oncol Clinics North America, 2006.) using as a basic model, a well‐defined effector T cell population in a series of studies which we intended to implement using the clinical and translational infrastructure we had already established. This led to a series of studies performed in rapid succession over the last 5 years:

1. Fludarabine Conditioning (Protocol 1796) PLOS One, 2009
a. Objective: Evaluate the influence of fludarabine lymphodepletion on the in vivo persistence of transferred CD8 T cell clones comparing an infusion given first without, and then 4 weeks later, a second infusion given following fludarabine conditioning.
b. Results:
i. Fludarabine conditioning led to predictable kinetics in lymphopenia and recovery
ii. T cell clones given following fludarabine conditioning experienced 3‐fold greater in vivo survival duration
iii. For a given CTL clone, its in vitro response to low‐dose IL‐15 was predictive of in vivo persistence
iv. Fludarabine led to an inversion in the CD4 Treg to CD4 Effector ratio suggesting increased resistance to fludarabine or faster recovery following depletion among Tregs.
c. Significance:
i. First demonstration of potential influence of fludarabine alone on T cell persistence
ii. A strategy for prospectively identifying CTL clones of potentially high replicative potential on a functional basis was demonstrated d. Future studies: While fludarabine conditioning did lead to extended in vivo persistence, clinical responses were modest and its differential effect on Treg vs. T effectors was concerning. Another approach to host conditioning was warranted.
iii. First demonstration that fludarabine can lead to inversion of CD4 Treg/Effector ratio upon homeostatic lymphocyte recovery. Preliminary studies suggest increased fludarabine-resistance to apoptosis among Treg cells.

2. Cyclophosphamide Conditioning Protocol 2140
a. Objective: Evaluate the influence of high dose cyclophosphamide lymphodepletion on the in vivo persistence of transferred CD8 T cell clones. Patients received low‐dose IL‐2 x 14 days post‐infusion.
b. Results:
i. Marked elevations in serum levels of homeostatic cytokines IL‐7 and IL‐15 were induced following cyclophosphamide conditioning
ii. In vivo persistence of transferred CTL clones up to 10+ months was observed
iii. > 60% response rate, including one durable CR was observed in first 5 patients. iv. Immune‐related adverse events (skin rash, cytokine release syndrome, fever) was directly correlated with clinical response
c. Significance:
i. Most clinically responsive regimen to date
ii. No serious unexpected toxicities
iii. First study demonstrating minimal requirement for prolonged in vivo persistence of transferred CTL clones. Future studies will evaluate the same conditiong regimen (high dose cyclophosphamide pre-infusion) but with high‐dose IL‐2 post‐infusion.

3. Adoptive CD4 T cell therapy (Protocol 1585, R01). New Engl J of Med, 2008
a. Objective: Evaluate the feasibility, safety and in vivo persistence of adoptively transferred antigen‐specific CD4 T cell clones.
b. Results:
i. Up to 1010 CD4 T cell clones can be administered safely.
ii. Adoptively transferred CD4 T cell clones can persist in vivo for several months in vivo in the absence of pre‐infusion conditioning or post‐infusion immunomodulation.
iii. Induction of non‐targeted endogenous responses to melanoma‐associated antigens can be achieved by targeting a single tumor-associated antigenic epitope. i.e. antigen‐spreading
iv. CD4 T cell therapy can mediate complete and partial responses in patients with refractory metastatic melanoma
c. Significance:
i. First‐in‐human study using antigen‐specific CD4 T cell clones.
ii. Complete response using T cell clones alone (no conditioning or immunomodulation)
The results of these 3 clinical trials represent our stepwise approach to addressing the requirements for successful immunotherapy. On the basis of these results we are currently enrolling patients on a clinical trial to evaluate the influence of CD4 T cell clones on concurrently administration CD8 T cell clones, (Protocol 2179) and, in a separate dose escalation study, the addition of cyclophosphamide conditioning to adoptive CD4 T cell therapy.

To avoid potential toxicities associated with high-dose IL-2 and optimize regulatory T cell depletion in vivo, the availability of novel clinical reagents such as diptheria-conjugated IL-2 (Ontak or Denileukin diftitox) provides us the unique opportunity to evaluate the influence of selectively depleting CD25hi regulatory T cells in vivo pre-infusion. A clinical trial incorporating the use of this reagent is opened in January 2010 (Protocol 2271)

In the context of combinational therapy, the addition of immunomodulatory reagents, such as checkpoint inhibitors (i.e. anti‐CTLA4) which lifts the regulatory `brake' on autoimmune responses may augment the function and survival of adoptively transferred T cells as well as lower the threshold to induction of endogenous T cell responses to tumor‐associated self antigens. This trial opened in 2009 (Protocol 2225).

These clinical trials are braced by the resources of the Immune Monitoring Laboratory, a Shared Resources facility, I established 6 years ago through CCSG funding. While I supervise the overall direction of this resource, the daily operational aspects are handled by a staff scientist and two technicians who provide comprehensive panel of leading edge immunologic assays for supporting primarily in‐house clinical trials, such as the adoptive therapy program but also, nonclinical, basic and murine studies, as well as, industry contracts.

It has become apparent from the work of others in the field and our own studies that the intrinsic properties of the T cell, in particular its replicative capacity and potential to persist as memory cells can influence its anti‐tumor efficacy.

We discovered that the recently cloned lymphokine, IL‐21, was unique in its capacity to elicit not only a ten‐fold greater frequency of antigen‐specific CTL when T cells are exposed to it during initial stimulation but also yield a population of CTL with markedly greater avidity and expansion potential. IL‐21 exposure led to the generation of a helper-independent CTL effectors with a unique memory phenotype: CD45RO+, CCR7‐, CD28+, and was capable of autocrine IL‐2 production following antigen stimulation (J Immunol, 2005). We postulate that strategies to enrich for a highly replicative tumor-antigen-specific T cell population will be instrumental in establishing T cell memory in vivo and durable clincial responses. It will also obviate the requirement to expand T cells to very large numbers (> 10^10), currently a potential hurdle to more widespread application of adoptive cellular therapy as a treatment modality.

We further demonstrated that this effect was due in part to the IL‐21‐mediated suppression/elimination of regulatory T cells and when used in combination with CD25 depletion led to virtual complete eradication of FoxP3+ T reg cells in vitro (not achievable currently by any previous non‐destructive means) and a > hundred‐fold increase in the frequency and absolute numbers of tumor‐associated antigen‐specific T cells (Blood, 2009).

We have also undertaken another tact for enhancing the anti‐tumor T cell response involving the use of superagonist altered peptide ligands as complementary means of augmenting the relatively rare frequency of T cell responses to tumor‐associated self antigens. Using this approach we have already demonstrated that in contrast to the existing dogma, individual superagonist APLs of the same epitope induce different outcomes among different patients and that a panel of superagonist APLs may be required to elicit a robust antitumor, antigen‐specific T cell response. The results of these studies have significant implications for induction of antigen‐specific T cells for adoptive therapy, but also for vaccination strategies.

Although our current clinical model was established and continues to be melanoma, we sought to evaluate the use of adoptive cellular therapy for the treatment of other malignances such as ovarian cancer, sarcoma and pancreatic cancer.

Ultimately, we believe that combined modalities incorporating vaccination, adoptive cellular therapy and manipulation of the tumor immune environment will be required for effective anti-tumor response and lifelong memory.

Child-rearing (n=2)

1. Adoptive T cell therapy of melanoma
2. Adoptive T cell therapy of ovarian cancer
3. Murine model of tumor immunity (or lack thereof)
4. Novel methods in immune monitoring
5. Novel methods in isolation and expansion of tumor-specific T cells