The Community Resource in Targeted Therapies
Driving Knowledge. Empowering Change. Optimizing Outcomes.
ONCAlert | Upfront Therapy for mRCC

Modeling T-Cell Trafficking to Increase the Likelihood of Radiation-Induced Abscopal Effects

Jan Poleszczuk, PhD; Eduardo G. Moros, PhD, Mayer Fishman, MD, PhD; Rachel Walker, PhD; Julie Djeu, PhD; Jonathan D. Schoenfeld, MD; Steven Finkelstein, MD; and Heiko Enderling, PhD
Published Online: Jun 28,2017

Abstract

The combination of radiation and immunotherapy is currently enjoying unprecedented attention as a treatment strategy for patients with metastatic cancer. Clinical case studies and proof-of-principle clinical trials report on systemic, abscopal responses to the combination of focal irradiation and immunotherapy in patients who were progressing on immunotherapy alone.1 However, individualized treatment plans to optimally exploit the synergy between immunotherapy and radiotherapy remain elusive due to high intra- and inter-patient heterogeneity and a myriad of possible radiation fractionation protocols, immunotherapy agents, and scheduling options.

 

Integrated mathematical oncology provides tools that could dissect this complexity and contribute to the transition of synergistic radiation and immunotherapy and radiation-induced abscopal effects into the personalized medicine era. To this end, we need to develop tractable, quantitative models based on carefully selected cancer biology and immunology principles. Predictions of such models, calibrated with patient-specific clinical data, need to be validated in prospective clinical trials.
 

 

Introduction

 

As a first step toward developing quantitative models, we recently developed a mathematical framework to simulate the systemic dissemination of T cells activated in response to focal therapy.2 Model simulations suggest that metastatic sites within individual patients do not participate equally in immune surveillance and thus are likely to exhibit different systemic responses after local irradiation. We hypothesized that such a model could help identify patient-specific radiation treatment targets that have a high likelihood of inducing abscopal effects. Such targeted treatment strategies would be then worthy of validation in a prospective clinical trial. In a subsequent commentary, this model was critically discussed, with a focus on complex biology that was not incorporated in the model.3 Here we discuss the raised concerns in light of the purpose and applicability of the mathematical model. We echo the need for clinical validation.

 

Clinical immunotherapy trials, especially if combined with adjuvant focal cytotoxic therapies, have generated encouraging results, including evidence of clinical remissions.4,5 Radiation therapy can synergize with immunotherapies, such as anti–CTLA-4 antibodies or FMS-like tyrasine kinase 3 ligands to generate systemic responses outside the radiation field, as observed in a series of seminal studies by Demaria and Formenti.6-8 Immune activation after radiation-induced immunogenic cell death provides an explanation for the previously believed anecdotal abscopal effect. Clinical case reports of radiation-induced abscopal effects date back to the 1950s9 and are reported after other immune-activating local therapies such as thermotherapies, including hyperthermia, radiofrequency ablation, and cryotherapy.10-12

 

The FDA approval of multiple immunotherapeutic agents generates both a clinical need and a prime opportunity to explore and exploit radiation and immunotherapy synergy, particularly for patients with metastatic cancer. In a recent proof-of-principle clinical trial combining local radiotherapy and granulocyte-macrophage colony-stimulating factor, 11 of 41 patients exhibited an objective systemic (abscopal) response.1 However, which metastatic sites were irradiated remained a heuristic decision.

 

Identifying patient-specific treatment targets adds an additional layer of personalization based upon limited clinical and biological data available for decision making. Unraveling the complex, adaptive tumor–immune system interactions that determine a response to therapy—both locally in the primary tumor and systemically in metastatic disease—requires nonlinear understanding and analysis of the multifactorial dynamics that govern them. As identified by Demaria and Formenti, more basic and translational research is needed to decrease treatment outcome uncertainty associated with the biological complexity of these interactions.3 Such research should include: 1) best radiation technique and fractionation protocol to induce antitumor immunity, 2) sequencing of immune modulators, 3) radiation and immunotherapy sensitivity of different tumor types in primary or metastatic tissue environments, 4) possible lack of common expression of antigen(s) or neoantigens between the irradiated and unirradiated metastases, and 5) components of the local immune environment, such as the availability and infiltration of dendritic cells.

 

As a complementary and arguably synergistic methodology to in vitro and in vivo models, mathematical models may contribute to translating basic biological principles into clinical decision making.13-16 We argue that some of the above-listed uncertainties may be addressed with a quantitative modeling approach where actionable data are available. Tractable, quantitative models can be designed, based on fundamental principles of cancer biology and immunology, to isolate and investigate key mechanisms and relationships within inherently complex biological systems. At present, attempting to elucidate the mechanisms that govern patient response to treatment based on a vast wealth of biological knowledge, but only a minimal amount of clinically obtainable patient information, is a near-impossible task. Instead, the tools of mathematical oncology allow us to identify key players in biological processes by using readily obtainable clinical and experimental data and evaluating its role in the broader setting of patient response. Where such data-driven models demonstrate the ability to reproduce or predict actual clinically observed phenomena, the mechanism in question can be assessed in a prospective clinical trial.

 

Mathematical models of tumor-immune interactions and therapy17 may be informed, parameterized, and calibrated by the experimental and clinical data collected by Demaria and Formenti7,8 and others.18 Mathematical models, like all biological modeling systems including cell culture, tissue culture, or orthotopic xenograft models, to name but a few, are simplifications of tumors in the human body. Thus, they have inherent weaknesses. The only question of interest is whether the models are illuminating and useful.19 Poleszczuk and colleagues recently developed a mathematical model of T-cell trafficking in the circulatory system of the human body.2 Based on physiological blood ow properties and variation in metastatic tumor volumes and anatomic disease distribution, model simulations revealed that metastatic sites within a patient, when treated with localized therapy, may have different abilities to induce systemic responses. This suggests that the metastases to irradiate in each patient mark an additional, previously unappreciated confounding factor for abscopal responses in the clinic. The mathematical model could help identify tumor sites in individual patients that serve as immunogenic hubs. The model was not designed to, nor can it, predict abscopal effects; it may, however, help to identify patient-specific treatment targets that have the highest potential to mediate an abscopal effect.

 

In a recent commentary to Poleszczuk’s study2, Demaria and Formenti criticize the simplicity of this modeling approach, as it fails to incorporate a myriad of complex biological processes.3 One major argument is that the importance of activated T-cell dissemination patterns in triggering the abscopal effect is dominated by other factors, such as the availability of dendritic cells. Although a larger relative
importance of these other factors is undoubtedly possible, this statement lacks conclusive support. Without such backing, the entropy of activated T-cell distribution among metastatic sites could still be the apparent driver of biological reactions in unirradiated metastases, independent of all other factors, despite their relevance to the question at hand. We agree that the model needs clinical validation with
both retrospective and prospective data to investigate the generated biological hypotheses and to determine clinical applicability. There are encouraging clinical results suggesting that anatomical distribution of metastases could impact the likelihood of response to checkpoint blockade. In a recent phase II multicenter clinical trial of atezolizumab (Tecentriq) in metastatic urothelial carcinoma, patients with liver or visceral metastases had a lower objective response rate compared with patients without.20

 

Predicting T-cell traf cking based on biological and physiological mechanisms, while certainly only 1 of many biological processes in abscopal responses, is exclusively calculable and predictable from routinely collected noninvasive imaging, independent of the primary tumor site. It remains currently infeasible to access the mutation load or immune in ltration in all metastatic sites in order to profile T-cell repertoires and to analyze intra- and inter-metastases antigen heterogeneity prior to focal radiation. Despite recent radiomics advances21, current radiographic assessments cannot predict immunogenic cell death following prospective irradiation, neither in a single tumor nor as a multimetastases comparison. We will extend our mathematical framework2 beyond its current simplicity when corresponding biological and clinical data become available to inform the model.

 

We applaud Demaria and Formenti for the proposal to generate an annotated registry of abscopal responders to inform, calibrate, and validate mathematical frameworks to help improve the likelihood of triggering abscopal responses. To date, however, there are too few abscopal effect studies with coherent data to create a statistically meaningful database. In addition to suggested anecdotal case reports,22-24 an ideal patient cohort to populate such an annotated registry comprises patients accrued to a clinical trial who were selected based on unifying clinical features and received the same treatment and longitudinal analysis. It is of equal importance to include nonresponder data in such a registry, as model predictability must also be guaranteed for patients without abscopal responses. A concerted effort must be made to ensure high-value clinical trial data from accrued patients is shareable after trial conclusion. This may bring together clinicians as well as quantitative and life scientists to ultimately advance our understanding of the complex biology underlying abscopal effects and to inform the next generation of clinical trials with available tools to understand and optimally apply synergistic treatment.

 
 
References:
  1. Golden EB, Chhabra A, Chachoua A, et al. Local radiotherapy and granulocyte-macrophage colony-stimulating factor to generate abscopal responses in patients with metastatic solid tumours: a proof-of-principle trial. Lancet Oncol. 2015 Jul;16(7):795–803.
  2. Poleszczuk J, Luddy KA, Prokopiou S, et al. Abscopal bene ts of localized radiotherapy depend on activated T cell tra cking and distribution between metastatic lesions. Cancer Res. 2016;76(5):1009–1018.
  3. Demaria S, Formenti SC. Can abscopal e ects of local radiotherapy be predicted by modeling T cell tra cking? J Immunother Cancer. 2016;4:29. doi: 1 0 . 11 8 6 / s 4 0 4 2 5 - 0 1 6 - 0 1 3 3 - 1 .
  4. Finkelstein SE, Iclozan C, Bui MM, et al. Combination of external beam radiotherapy (EBRT) with intratumoral injection of dendritic cells as neo- adjuvant treatment of high-risk soft tissue sarcoma patients. Int J Radiat Oncol Biol Phys. 2012;82(2):924–932.
  5. Crittenden M, Kohrt H2, Levy R. Current clinical trials testing combinations of immunotherapy and radiation. Semin Radiat Oncol. 2015;25(1):54-64. doi: 10.1016/j.semradonc.2014.07.003.
  6. Pilones KA, Vanpouille-Box C, Demaria S. Combination of radiotherapy and immune checkpoint inhibitors. Semin Radiat Oncol. 2015;25(1):28–33.
  7. Demaria S, Ng B, Devitt ML, et al. Ionizing radiation inhibition of distant untreated tumors (abscopal e ect) is immune mediated. Int J Radiat Oncol Biol Phys. 2004;58(3):862-870.
  8. Dewan MZ, Galloway AE, Kawashima N, et al. Fractionated but not single- dose radiotherapy induces an immune-mediated abscopal e ect when combined with anti-CTLA-4 antibody. Clin Cancer Res. 2009;15(17):5379- 5388. doi: 10.1158/1078-0432.CCR-09-0265.
  9. Abuodeh Y, Venkat P, Kim S. Systematic review of case reports on the abscopal e ect. Curr Probl Cancer. 2016;40(1):25-37. doi: 10.1016/j. currproblcancer.2015.10.001.
  10. Toraya-Brown S, Fiering S. Local tumour hyperthermia as immunotherapy for metastatic cancer. Int J Hyperthermia. 2014;30(8):531-539. doi: 10.3109/02656736.2014.968640.
  11. O’Brien MA, Power DG, Clover AJP, Bird B, Soden DM, Forde PF. Local tumour ablative therapies: opportunities for maximising immune engagement and activation. Biochim Biophys Acta. 2014;1846(2):510-523. doi: 10.1016/j. bbcan.2014.09.005.
  12. SharsteinG,KaufmannY,HenningsL,etal.Conductiveinterstitial thermal therapy (CITT) inhibits recurrence and metastasis in rabbit VX2 carcinoma model. Int J Hyperthermia. 2009;25(6):446-454. doi: 10.1080/02656730903013618.
  13. Anderson ARA, Quaranta V. Integrative mathematical oncology. Nat Rev Cancer. 2008;8(3):227-234. doi: 10.1038/nrc2329.
  14. Rockne R, Rockhill JK, Mrugala M, et al. Predicting the e cacy of radiotherapy in individual glioblastoma patients in vivo: a mathematical modeling approach. Phys Med Biol. 2010;55(12):3271-3285. doi: 10.1088/0031- 9155/55/12/001.
  15. Yankeelov TE, Atuegwu N, Hormuth D, et al. Clinically relevant modeling of tumor growth and treatment response. Sci Transl Med. 2013;5(187):187ps9. doi: 10.1126/scitranslmed.3005686.
  16. Prokopiou S, Moros EG, Poleszczuk J, et al. A proliferation saturation index to predict radiation response and personalize radiotherapy fractionation. Radiat Oncol. 2015;10:159. doi: 10.1186/s13014-015-0465-x.
  17. Walker R, Enderling H. From concept to clinic: mathematically informed immunotherapy. Curr Probl Cancer. 2016;40(1):68-83. doi: 10.1016/j. currproblcancer.2015.10.004.
  18. Chandra RA, Wilhite TJ, Balboni TA, et al. A systematic evaluation of abscopal responses following radiotherapy in patients with metastatic melanoma treated with ipilimumab. Oncoimmunology. 2015;4(11):e1046028.
  19. Box GEP. Robustness in the Strategy of Scienti c Model Building. Robustness in Statistics. Elsevier; 1979. p. 201–236.
  20. Rosenberg JE, Ho man-Censits J, Powles T, et al. Atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum-based chemotherapy: a single- arm, multicentre, phase 2 trial. Lancet. 2016;387(10031):1909-1920. doi: 10.1016/S0140 - 6736(16)00561- 4.
  21. Gillies RJ, Kinahan PE, Hricak H. Radiomics: images are more than pictures, they are data. Radiology. 2016;278(2):563–577. doi: 10.1148/ radiol.2015151169.
  22. Postow MA, Callahan MK, Barker CA, et al. Immunologic correlates of the abscopal e ect in a patient with melanoma. N Engl J Med. 2012;366(10):925- 931. doi: 10.1056/NEJMoa1112824.
  23. Golden EB, Demaria S, Schi PB, Chachoua A, Formenti SC. An abscopal response to radiation and ipilimumab in a patient with metastatic non-small cell lung cancer. Cancer Immunol Res. 2013;1(6):365-372. doi: 10.1158/2326- 6066.CIR-13-0115.
  24. Schoenfeld JD, Mahadevan A, Floyd SR, et al. Ipilmumab and cranial radiation in metastatic melanoma patients: a case series and review. J Immunother Cancer. 2015;3:50. doi: 10.1186/s40425-015-0095-8.



Clinical Articles

Modeling T-Cell Trafficking to Increase the Likelihood of Radiation-Induced Abscopal Effects
Copyright © TargetedOnc 2018 Intellisphere, LLC. All Rights Reserved.