**5. RT and immunotherapy combination modalities**

### **5.1 Pre-clinical models**

202 Prostate Cancer – Diagnostic and Therapeutic Advances

probably due to loss of membrane-bound TGF-β (Cao et al., 2009). Tregs were especially sensitive to low-dose radiation compared to effector T cells (Cao et al., 2011). Another study, in TRAMP mice, had the opposite findings: Treg cell numbers increased in immune organs after local or whole body irradiation without changes to their functional activity (Kachikwu et al., 2010), indicating relative resistance to 0-20 Gy radiation. It remains to be seen what

We showed recently that lymphopenia following prostate and pelvic RT causes preferential death of naïve or unstimulated T-cells (Tabi et al., 2010). Elevated frequencies of Treg cells were observed in the circulation following 44 Gy radiation in 20 fractions to the pelvic nodes and 55 Gy to the prostate (Fig.2). T cell proliferative function was also impaired (Tabi et al., 2010) but it was restored in vitro with exogenous IL-2 without increasing Treg frequencies (Fig. 2). This indicates that IL-2 maybe used to support T cell function after patients

> IL-2 (U/ml) 0 20 100

Fig. 2. IL-2 response of Treg cells from the peripheral blood of PCa patients undergoing standard RT. Frequencies of CD4+CD25+Foxp3+ T cells (see gate in insert) were measured before (RT0), immediately after radical radiation in 20 fractions (RT20) and 4 weeks after the last fraction (pRT4). Means and SD of triplicates are shown from a representative patient. Frequencies of Tregs were elevated at RT20, but returned to pre-radiation (RT0) level at pRT4. Unlike at RT0 or RT20, exogenous IL-2, added to the cells in vitro, did not increase

Most importantly, we identified novel TAA-specific T-cell responses post-RT (Tabi et al., 2010), which were not present before RT. Similar findings were observed by others (Nesslinger et al., 2007; Schaue et al., 2008), indicating the ability of radiation to shift the

balance between tumour-specific regulatory and effector immune mechanisms.

RT0 RT20 pRT4

happens to Tregs in situ during RT of PCa.

completed standard RT.

% CD4+CD25+Foxp3+ cells

0

Treg frequencies at pRT4.

1

2

3

4

5

6

7

The use of IL-2 as a monotherapy in cancer has been extensively researched but due to issues with toxicity its clinical use has been limited. In the murine renal adenocarcinoma model, IR was found to augment the response of pulmonary metastases to IL-2 therapy (Younes et al., 1995). Following IR to one lung, plus systemic IL-2 treatment, a reduction in tumour size was observed in both lungs. The effect is dose-dependent and immunohistological studies show significant infiltration of T cells and macrophages at the tumour site. IL-2 is capable of rescuing T cells from IR-induced apoptosis and restores T cell proliferation after RT (Tabi et al., 2010). Its use in combination with RT may minimise the immunosuppressive effects of RT and enhance tumour cell killing via T cells (Mor & Cohen, 1996). In PCa, the combination of IL-2 and radiation in a mouse bone metastases model demonstrated a ~50% inhibition of tumour growth (Hillman et al., 2003). There was a greater degree of tumour destruction in IL-2-treated irradiated tumours than in irradiated tumours alone and the histology revealed increased fibrosis and elevated numbers of infiltrating inflammatory immune cells.

Antitumour effects were also observed in a model utilising IL-12 and RT. IL-12 is secreted by mature DC and macrophages and required for IFNγ and TNFα production from T cells and mediates a Th1-type immune response. Adenovirus-derived IL-12 plus RT significantly increased local antitumour and systemic antimetastatic effects in a preclinical model of metastatic PCa when compared to either treatment alone (Fujita et al., 2008). This antimetastatic activity is due to the antitumoural activities of natural killer (NK) cells. These results were also observed in the 4T1 mammary carcinoma. The combination of RT and an adenoviral vector encoding IL-12 and the co-stimulatory molecule CD80 resulted in a significant reduction in tumour growth (Lohr et al., 2000). The antitumour effect observed in the combination therapy group was far superior if the IL-12 and CD80-expressing adenovirus was administered after the final fraction of radiation.

Further cytokine studies have evaluated the combined effect of IL-3 and RT. IL-3 differentiates haematopoietic stem cells into myeloid progenitor cells and stimulates the proliferation of myeloid-derived cells such as DC and monocytes. In mouse models of fibrosarcoma and PCa, IL-3 was found to increase the tumour response to radiation. Combining adenoviral-IL-3 and radiation in the TRAMP-C1 mouse prostate model caused significant delays in tumour growth. Further reports indicated that adenoviral-IL-3 plus radiation enhanced IFNγ-producing CD4+ and CD8+ T cell responses in the spleen (Oh et al., 2004). This shifted the immune response to a Th1–type response from a suppressive Th2 type response (Tsai et al., 2006).

The combination of the pro-inflammatory cytokine TNFα and RT in mouse mammary carcinoma delayed tumour growth at a greater extent than either treatment alone (Nishiguchi et al., 1990). Similar synergistic effects have also been observed in mouse melanoma, lung adenocarinoma and brain tumours. The combined effects were attributed to increased recruitment and enhanced activation of lymphocytes and neutrophils (Gridley et al., 1996; Gridley et al., 2002; Gridley et al., 2000; Jin et al., 2005; Li et al., 1998).

Adoptive-cell-transfer (ACT) therapy is the passive transfer of tumour-specific T cells that have been expanded ex vivo. Local tumour irradiation can enhance the therapeutic efficacy of ACT therapy (Teitz-Tennenbaum et al., 2009). Combination of RT with ACT of carcinoembryonic antigen (CEA)-specific CD8+ T cells in a mouse colon carcinoma

Combination of Immunotherapy & Radiotherapy for the Treatment of Prostate Cancer 205

but the hazard ratio for death was 0.56 (95% CI 0.37 to 0.85) in the PSA-TRICOM arm and

The initial results of the phase II trial in combination with radiation were highly encouraging. Thirty patients were randomised in a 2:1 ratio to receive vaccine plus EBRT or EBRT alone. In this trial the vaccine consisted of a priming vaccine with rV-PSA plus rV-B7.1 followed by monthly booster vaccines with rFP-PSA. The immunological adjuvants used were GMCSF and high dose IL-2. PSA-specific T cell responses generated prior to RT were not adversely affected by RT, confirming our observation (Tabi et al., 2010). In total, 13 of the 17 patients in the combination arm had increases in PSA-specific T cells and epitope spreading to 4 other prostate cancer TAA (PSMA, PAP, PSCA and MUC-1) was noted (Gulley et al., 2005), possibly due to cross-presentation of a mix of TAA from dying tumour cells by DC (Obeid et al., 2007). There was some IL-2 toxicity, which was reduced in a singlearm follow-up study using lower doses but longer durations of IL-2; immunological effects

There is one other reported ongoing work of immunotherapy-radiotherapy combination with intraprostatic injections of autologous DC. The first report confirms the safety of this approach in 5 HLA-A2+ patients with high risk, localised disease, also treated with ADT, EBRT to 45 Gy and LDR brachytherapy. Autologous intraprostatic DC injections were given at four timepoints during EBRT. Measurable, induced increases in TAA-specific T cell frequencies in peripheral blood using ELISPOT were observed in some patients. The pattern of distribution of CD8+ cells in tissue was consistent with PCa TAA-targeting, rather than

There are no other published studies in PCa patients of EBRT in combination with cytokines, antibodies, immune modulators or immunologically relevant gene therapy. However, there are a few prostate cancer clinical trials combining RT and immunotherapy currently recruiting according to the ClinicalTrials.gov website, such as anti-CTLA-4 (Ipilimumab) antibody therapy in castration-resistant prostate cancer following RT (Phase III trial, Bristol-Myers Squibb; NCT00861614) and treatment with anti-OX40 and cyclophosphamide in combination with RT in metastatic prostate cancer (Providence Health

Clinical trial design, investigating the benefit of immunotherapy in addition to EBRT is challenging in PCa, as long-term tumour control outcome is already good in both localised (assuming dose escalated image-guided IMRT) and locally advanced disease (assuming combination of long term ADT and EBRT). Phase III trials, adequately powered to show a clinically relevant improvement, would need to address biochemical relapse-free survival or overall survival, both of which require prolonged follow-up of many hundreds and perhaps thousands of patients. Before such trials were undertaken, it is important to optimise the immunotherapy-RT schedules. Reliable biomarkers of treatment-efficacy are needed and this is difficult, especially if neoadjuvant or adjuvant ADT is used, as changes in PSAkinetics become redundant in these patients. Therefore, development of reliable immunological biomarkers is crucial. We believe that the presence of systemic TAA-specific T cell responses is likely to be the most reliable and easily detectable indicator of a sustainable immunological effect. It may also assist patient selection for optimised treatment

of those patients who are most likely to benefit from combination therapies.

the treatment was generally well tolerated.

were equivalent (Gulley et al., 2005; Lechleider et al., 2008).

non-specific organ infiltration (Finkelstein et al., 2011 ).

& Services, Oregon USA, Phase I/II trial; NCT01303705).

**5.3 Designs of combination clinical trials** 

demonstrated increased tumour rejection, that could be attributed to the upregulation of the death receptor Fas on the surface of irradiated tumour cells (Chakraborty et al., 2003).

The anti-tumour effect can be further enhanced by intratumoural administration of DC. In a murine metastatic melanoma model reduction in the size of tumour and extent of spontaneous metastasis and prolonged survival were observed after irradiation and intratumoural DC administration. This was associated with an increase in proliferation, accumulation and cytokine production of CD4+ cells. Similar results were observed in DC plus irraditation in melanoma and sarcoma models (J. Huang et al., 2007). Kjaergaard et al. (2005) reported a method of fusing DC with tumour cells via an electric field resulting in a TAA-DC primed vaccine. These DC/tumour cell fusions induce a potent immune response in combination with local cranial RT in mouse glioma. Both CD4+ and CD8+ T cells infiltrate the tumours, leading to complete tumour regression. Tumour rejection was also observed after subsequent tumour challenge, indicating the presence of immunological memory.

Vaccines containing either modified tumour cells that are more immunogenic than the "native" tumour cells or TAA-vaccines have been tested extensively in preclinical models. Combination of cytokine-producing vaccines with local RT in mouse glioma demonstrated that IL-4 and GM-CSF vaccines alone were capable of curing 20-40% of mice but in combination with local RT 80-100% of the mice were cured. The brain tumours were heavily infiltrated with CD4+ T cells (Lumniczky et al., 2002). The increased anti-tumour effect of GM-CSF and RT was also demonstrated in mouse glioma using a vaccine which contained GM-CSF-secreting tumour cells (Newcomb et al., 2006).

Strategies using RT in combination with monoclonal antibodies that are specific for TAA are now commonly used in clinical oncology (reviewed by Drake, 2010). In a mouse lung cancer model, monoclonal antibody to OX40 (a secondary co-stimulatory molecule expressed on activated CD4+ and CD8+ T cells) and RT resulted in a synergistic effect on survival compared to either treatment alone (Yokouchi et al., 2008). The effect was CD8+ T cell dependent. Antibody-based immunotherapy strategies aiming to neutralise molecules implied in immune tolerance have also been examined. Antibodies for the cytotoxic T lymphocyte antigen (CTLA)-4 (a CD28-superfamily molecule causing T cell functional inhibition) have been shown to induce effective anti-tumour responses via lowering the threshold of tumour-specific T cell activation (reviewed by Drake, 2010). Based on the preclinical findings, CTLA-4 inhibition by the antibody Ipilimumab is now an FDA approved treatment of metastatic melanoma (Chambers et al., 2001). Its combination with RT is being tested in animal models (Dewan et al., 2009) and in clinical trials (see next section).

#### **5.2 Clinical trials**

There is huge potential for augmentation of the radiation response with the use of immunotherapy but as yet there are only a few clinical trials published. These were carried out in PCa patients at the National Cancer Institute, using a recombinant viral vaccine consisting of recombinant vaccinia virus (rV) encoding PSA, admixed with rV encoding the co-stimulatory molecule B7.1, followed by booster vaccinations with recombinant fowlpox (rFP) vector expressing PSA prior to RT. The product has been further developed and is presently marketed as Prostvac® which encodes ICAM-1 and LFA-3 in addition to B7.1 (PSA-TRICOM). This agent has been shown to improve median overall survival from 16.6 months to 25.1 months in a phase II multi-centre randomized controlled trial in 125 men with asymptomatic or minimally symptomatic metastatic castrate refractory prostate cancer (Kantoff et al., 2010). There was similar progression-free survival in the two arms of the trial,

demonstrated increased tumour rejection, that could be attributed to the upregulation of the death receptor Fas on the surface of irradiated tumour cells (Chakraborty et al., 2003). The anti-tumour effect can be further enhanced by intratumoural administration of DC. In a murine metastatic melanoma model reduction in the size of tumour and extent of spontaneous metastasis and prolonged survival were observed after irradiation and intratumoural DC administration. This was associated with an increase in proliferation, accumulation and cytokine production of CD4+ cells. Similar results were observed in DC plus irraditation in melanoma and sarcoma models (J. Huang et al., 2007). Kjaergaard et al. (2005) reported a method of fusing DC with tumour cells via an electric field resulting in a TAA-DC primed vaccine. These DC/tumour cell fusions induce a potent immune response in combination with local cranial RT in mouse glioma. Both CD4+ and CD8+ T cells infiltrate the tumours, leading to complete tumour regression. Tumour rejection was also observed after subsequent tumour challenge, indicating the presence of immunological memory. Vaccines containing either modified tumour cells that are more immunogenic than the "native" tumour cells or TAA-vaccines have been tested extensively in preclinical models. Combination of cytokine-producing vaccines with local RT in mouse glioma demonstrated that IL-4 and GM-CSF vaccines alone were capable of curing 20-40% of mice but in combination with local RT 80-100% of the mice were cured. The brain tumours were heavily infiltrated with CD4+ T cells (Lumniczky et al., 2002). The increased anti-tumour effect of GM-CSF and RT was also demonstrated in mouse glioma using a vaccine which contained

Strategies using RT in combination with monoclonal antibodies that are specific for TAA are now commonly used in clinical oncology (reviewed by Drake, 2010). In a mouse lung cancer model, monoclonal antibody to OX40 (a secondary co-stimulatory molecule expressed on activated CD4+ and CD8+ T cells) and RT resulted in a synergistic effect on survival compared to either treatment alone (Yokouchi et al., 2008). The effect was CD8+ T cell dependent. Antibody-based immunotherapy strategies aiming to neutralise molecules implied in immune tolerance have also been examined. Antibodies for the cytotoxic T lymphocyte antigen (CTLA)-4 (a CD28-superfamily molecule causing T cell functional inhibition) have been shown to induce effective anti-tumour responses via lowering the threshold of tumour-specific T cell activation (reviewed by Drake, 2010). Based on the preclinical findings, CTLA-4 inhibition by the antibody Ipilimumab is now an FDA approved treatment of metastatic melanoma (Chambers et al., 2001). Its combination with RT is being tested in animal models

There is huge potential for augmentation of the radiation response with the use of immunotherapy but as yet there are only a few clinical trials published. These were carried out in PCa patients at the National Cancer Institute, using a recombinant viral vaccine consisting of recombinant vaccinia virus (rV) encoding PSA, admixed with rV encoding the co-stimulatory molecule B7.1, followed by booster vaccinations with recombinant fowlpox (rFP) vector expressing PSA prior to RT. The product has been further developed and is presently marketed as Prostvac® which encodes ICAM-1 and LFA-3 in addition to B7.1 (PSA-TRICOM). This agent has been shown to improve median overall survival from 16.6 months to 25.1 months in a phase II multi-centre randomized controlled trial in 125 men with asymptomatic or minimally symptomatic metastatic castrate refractory prostate cancer (Kantoff et al., 2010). There was similar progression-free survival in the two arms of the trial,

GM-CSF-secreting tumour cells (Newcomb et al., 2006).

(Dewan et al., 2009) and in clinical trials (see next section).

**5.2 Clinical trials** 

but the hazard ratio for death was 0.56 (95% CI 0.37 to 0.85) in the PSA-TRICOM arm and the treatment was generally well tolerated.

The initial results of the phase II trial in combination with radiation were highly encouraging. Thirty patients were randomised in a 2:1 ratio to receive vaccine plus EBRT or EBRT alone. In this trial the vaccine consisted of a priming vaccine with rV-PSA plus rV-B7.1 followed by monthly booster vaccines with rFP-PSA. The immunological adjuvants used were GMCSF and high dose IL-2. PSA-specific T cell responses generated prior to RT were not adversely affected by RT, confirming our observation (Tabi et al., 2010). In total, 13 of the 17 patients in the combination arm had increases in PSA-specific T cells and epitope spreading to 4 other prostate cancer TAA (PSMA, PAP, PSCA and MUC-1) was noted (Gulley et al., 2005), possibly due to cross-presentation of a mix of TAA from dying tumour cells by DC (Obeid et al., 2007). There was some IL-2 toxicity, which was reduced in a singlearm follow-up study using lower doses but longer durations of IL-2; immunological effects were equivalent (Gulley et al., 2005; Lechleider et al., 2008).

There is one other reported ongoing work of immunotherapy-radiotherapy combination with intraprostatic injections of autologous DC. The first report confirms the safety of this approach in 5 HLA-A2+ patients with high risk, localised disease, also treated with ADT, EBRT to 45 Gy and LDR brachytherapy. Autologous intraprostatic DC injections were given at four timepoints during EBRT. Measurable, induced increases in TAA-specific T cell frequencies in peripheral blood using ELISPOT were observed in some patients. The pattern of distribution of CD8+ cells in tissue was consistent with PCa TAA-targeting, rather than non-specific organ infiltration (Finkelstein et al., 2011 ).

There are no other published studies in PCa patients of EBRT in combination with cytokines, antibodies, immune modulators or immunologically relevant gene therapy. However, there are a few prostate cancer clinical trials combining RT and immunotherapy currently recruiting according to the ClinicalTrials.gov website, such as anti-CTLA-4 (Ipilimumab) antibody therapy in castration-resistant prostate cancer following RT (Phase III trial, Bristol-Myers Squibb; NCT00861614) and treatment with anti-OX40 and cyclophosphamide in combination with RT in metastatic prostate cancer (Providence Health & Services, Oregon USA, Phase I/II trial; NCT01303705).

#### **5.3 Designs of combination clinical trials**

Clinical trial design, investigating the benefit of immunotherapy in addition to EBRT is challenging in PCa, as long-term tumour control outcome is already good in both localised (assuming dose escalated image-guided IMRT) and locally advanced disease (assuming combination of long term ADT and EBRT). Phase III trials, adequately powered to show a clinically relevant improvement, would need to address biochemical relapse-free survival or overall survival, both of which require prolonged follow-up of many hundreds and perhaps thousands of patients. Before such trials were undertaken, it is important to optimise the immunotherapy-RT schedules. Reliable biomarkers of treatment-efficacy are needed and this is difficult, especially if neoadjuvant or adjuvant ADT is used, as changes in PSAkinetics become redundant in these patients. Therefore, development of reliable immunological biomarkers is crucial. We believe that the presence of systemic TAA-specific T cell responses is likely to be the most reliable and easily detectable indicator of a sustainable immunological effect. It may also assist patient selection for optimised treatment of those patients who are most likely to benefit from combination therapies.

Combination of Immunotherapy & Radiotherapy for the Treatment of Prostate Cancer 207

Baluna, R.G., Eng, T.Y. & Thomas, C.R. (2006). Adhesion molecules in radiotherapy. *Radiat* 

Begley, L., Monteleon, C., Shah, R.B., Macdonald, J.W. & Macoska, J.A. (2005). CXCL12

Bolla, M., Van Poppel, H. & Collette, L. (2007). [Preliminary results for EORTC trial 22911:

Bronte, V., Kasic, T., Gri, G., Gallana, K., Borsellino, G., Marigo, I., Battistini, L., Iafrate, M.,

Cao, M., Cabrera, R., Xu, Y., Liu, C. & Nelson, D. (2009). Gamma irradiation alters the

Cao, M., Cabrera, R., Xu, Y., Liu, C. & Nelson, D. (2011). Different radiosensitivity of

Carmeliet, P. & Jain, R.K. (2000). Angiogenesis in cancer and other diseases. *Nature* 407,

Chakraborty, M., Abrams, S.I., Camphausen, K., Liu, K., Scott, T., Coleman, C.N. & Hodge,

Chen, Y., Scanlan, M., Sahin, U., Türeci, O., Gure, A., Tsang, S., Williamson, B., Stockert, E.,

Chopra, D.P., Menard, R.E., Januszewski, J. & Mattingly, R.R. (2004). TNF-alpha-mediated

Clayton, A., Al-Taei, S., Webber, J., Mason, M.D. & Tabi, Z. (2011). Cancer exosomes express

Coen, J. & Zietman, A. (2009). Proton radiation for localized prostate cancer. *Nat Rev Urol* 6,

Dewan, M., Galloway, A., Kawashima, N., Dewyngaert, J., Babb, J., Formenti, S. & Demaria,

with a high risk of progression]. *Cancer Radiother* 11, 6-7, (Nov), 363-369. Bouchelouche, K., Tagawa, S.T., Goldsmith, S.J., Turkbey, B., Capala, J. & Choyke, P. (2011).

epithelial proliferation in vitro. *Aging Cell* 4, 6, (Dec), 291-298.

overexpression and secretion by aging fibroblasts enhance human prostate

radical prostatectomy followed by postoperative radiotherapy in prostate cancers

PET/CT imaging and radioimmunotherapy of prostate cancer. *Semin Nucl Med* 41,

Prayer-Galetti, T., Pagano, F. & Viola, A. (2005). Boosting antitumor responses of T lymphocytes infiltrating human prostate cancers. *J Exp Med* 201, 8, 1257-1268. Burnette, B.C., Liang, H., Lee, Y., Chlewicki, L., Khodarev, N.N., Weichselbaum, R.R., Fu,

Y.X. & Auh, S.L. (2011). The efficacy of radiotherapy relies upon induction of type I interferon -dependent innate and adaptive immunity. *Cancer Res* 71, 7, (Apr 1),

phenotype and function of CD4+CD25+ regulatory T cells. *Cell Biol Int* 33, 5, (May),

CD4+CD25+ regulatory T cells and effector T cells to low dose gamma irradiation

J.W. (2003). Irradiation of tumor cells up-regulates Fas and enhances CTL lytic activity and CTL adoptive immunotherapy. *J Immunol* 170, 12, (Jun 15), 6338-6347. Chambers, C., Kuhns, M., Egen, J. & Allison, J. (2001). CTLA-4-mediated inhibition in

regulation of T cell responses: mechanisms and manipulation in tumor

Pfreundschuh, M. & Old, L. (1997). A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening. *Proc Natl Acad Sci USA* 

apoptosis in normal human prostate epithelial cells and tumor cell lines. *Cancer Lett* 

CD39 and CD73, which suppress T cells through adenosine production. *J Immunol* 

S. (2009). Fractionated but not single-dose radiotherapy induces an immunemediated abscopal effect when combined with anti-CTLA-4 antibody. *Clin Cancer* 

*Res* 166, 6, (Dec), 819-831.

1, (Jan), 29-44.

2488-2496.

565-571.

in vitro. *Int J Rad Biol* 87, 1, 71-80.

immunotherapy. *Ann Rev Immunol* 19, 565-594.

6801, (Sep 14), 249-257.

94,5,1914-1918.

6, 324-330.

203, 2, (Jan 20), 145-154.

187, 2, (July 15) 676-683.

*Res* 15, 17, 5379-5388.

The successful result of the IMPACT trial showing the survival advantage with Sipuleucel-T is expected to lead to a dramatic increase in the use of systemic immunotherapy for prostate cancer (Sonpavde et al., 2011). Presently there are only production facilities within the USA and the cost is likely to remain prohibitive for many healthcare systems. Radiation has great potential to augment the effect of systemic immunotherapy and we can expect many combination trials in the future. Clinical trial design in this setting will remain challenging as immunotherapy appears to improve overall survival but demonstrable tumour or biochemical responses are often delayed. Augmenting immunotherapy with radiation may improve overall advantage further.
