A tumor-in-host Dynamic Energy Budget (DEB)-based paradigm for preclinical to clinical translation: predictions of tumor-in-host growth dynamics
Elena Maria Tosca (1), Stefano Castellano (1), Davide Ronchi (1), Maurizio Rocchetti (2), Paolo Magni (1)
(1) Department of Electrical, Computer and Biomedical Engineering, University of Pavia, via Ferrata 5, Pavia, I-27100, Italy, (2) Consultant, Milano, Italy
Objectives: Despite its complexity, human tumor growth displays a relatively simple time curve. Human tumors were generally found to start from a single cell, to have a long subclinical period before reaching the detection level, to follow an S-shaped growth pattern during the measurable phase and, finally, to reach an upper limit (~1012-1013cells ≈1-10kg) at which the tumor burden become lethal. Irrespective of the growth rates, the clinical (and measurable) period is generally characterized by an exponential grow defined by a constant rate, λ (1). During this phase, in the absence of biological or molecular predictive markers, tumor volume doubling time (TVDT≈ln(2)/λ) is considered a non-invasive assay accommodating all factors that influence tumor dynamics (rates of proliferation, neo-angiogenesis, apoptosis and necrosis and immune system effects). TVDTs vary significantly between patients but less than between different cancer types. Information about growth dynamics and DTs are essential for important strategies in oncology such as screening programs, survival data analysis during clinical trials or estimation of risk period of late recurrences.
Recently we developed a Dynamic Energy Budget (DEB)-based model able to describe tumor-in-host dynamics in xenograft mice (2,3). The model predicts an S-shaped tumor growth profile in which the exponential phase is defined by a constant rate λ0 strictly linked to the tumor and host characteristics. Here, we propose a scaling strategy to translate the tumor-in-host growth from xenograft mice to humans.
Methods: Experimental: Preclinical xenograft studies performed on mice were considered. They involved a panel of tumor cell lines derived from diverse human cancer types: A2780 ovarian cancer, A375 and A2058 melanoma, MX-1, KPL4, CTG-033 and PDX-63 breast cancer, SNU-16 gastric cancer, HT-29, Colon-26, Colo-205, C26 and HTC-116 colon adeno/carcinoma, S2-013 and BX-PC3 pancreatic,HCC827 lung adenocarcinoma and DU145 prostate cancer cell line. TVDT estimates obtained from patients for the considered cancer types were collected from literature (1,4-11). Modeling: The nesting structure of the model, in which the host and tumor description were well distinct, allowed to perform the translation in proceeding steps. First, the host growth was scaled in absence of tumor. Scaling rules from preclinical species to human were defined based on body size and life-span using only weight profiles of tumor-free individuals. Secondly, the scaling rules previously identified were applied to scale the tumor-in-host growth from mice to humans.
Results: The proposed scaling approach allowed us to translate the tumor-in-host growth from mice to humans. First, for each preclinical cancer model the host-related, tumor-related and cachexia-related parameters were estimated on host body and tumor weights collected during xenograft studies. Then, scaling factors were identified on growth chart data of tumor-free individuals. Finally, the obtained estimates were used to scale the tumor-in-growth model. To validate our approach, model predicted and clinically estimated TVDTs were compared. For each cancer cell line, the obtained estimates were in good agreement with observations in patients affected by the correspondent cancer type. Model predicted and clinical TVDT [days] were 44 vs 75 for ovarian, 82 vs 94 for melanoma, 438 vs 282 for breast, 256 vs 303 for gastric, 96 vs 211for colon, 105 vs 146 for pancreatic cancer, 73 vs 67 for prostate and 261 vs 222 for lung adenocarcinoma. The absolute average fold error was always <1 with good correlation (r2>0.8) between predicted and observed mean TVDTs. Moreover, the scaled model was used to simulate human body weight dynamics during cancer progression. Model predicted body weight reductions were in line with cachectic state often observed in cancer patients (12,13). Model predicted a deep body weight loss when tumor reaches a volume usually considered as a lethal tumor burden.
Conclusions: In this work, a tumor-in-host paradigm for preclinical to clinical translation was proposed. Starting from model estimates obtained on xenograft studies, we were able to predict TVDTs in agreement to these observed in cancer patients. Moreover, obtained results confirmed the ability of the proposed tumor-in-host DEB-based approach to qualitative predict human tumor-in-host kinetics. This preliminary assessment of the model is a good starting point for further investigations focused on the translation of drug activity and cachexia dynamics.
References:
[1] Klein, C. A. (2009). Parallel progression of primary tumours and metastases. Nature Reviews Cancer, 9(4), 302-312.
[2] Tosca, E. M., Rocchetti, M., Pesenti, E., & Magni, P. (2020). A tumor-in-host DEB-based approach for modeling cachexia and bevacizumab resistance. Cancer Research, 80(4), 820-831
[3] Terranova, N., Tosca, E. M., Borella, E., Pesenti, E., Rocchetti, M., & Magni, P. (2018). Modeling tumor growth inhibition and toxicity outcome after administration of anticancer agents in xenograft mice: A Dynamic Energy Budget (DEB) approach. Journal of theoretical biology, 450, 1-14.
[4] Danesh, K., Durrett, R., Havrilesky, L. J., & Myers, E. (2012). A branching process model of ovarian cancer. Journal of theoretical biology, 314, 10-15.
[5] Carlson, J. A. (2003). Tumor doubling time of cutaneous melanoma and its metastasis. The American journal of dermatopathology, 25(4), 291-299.
[6] Förnvik, D., Lång, K., Andersson, I., Dustler, M., Borgquist, S., & Timberg, P. (2016). Estimates of breast cancer growth rate from mammograms and its relation to tumour characteristics. Radiation protection dosimetry, 169(1-4), 151-157.
[7] Haruma, K., Suzuki, T., Tsuda, T., Yoshihara, M., Sumii, K., & Kajiyama, G. (1991). Evaluation of tumor growth rate in patients with early gastric carcinoma of the elevated type. Gastrointestinal radiology, 16(1), 289-292.
[8] Burke, J. R., Brown, P., Quyn, A., Lambie, H., Tolan, D., & Sagar, P. (2020). Tumour growth rate of carcinoma of the colon and rectum: retrospective cohort study. BJS open, 4(6), 1200-1207.
[9] Hruban, R. H., Gaida, M. M., Thompson, E., Hong, S. M., Noë, M., Brosens, L. A., ... & Wood, L. D. (2019). Why is pancreatic cancer so deadly? the pathologist's view. The Journal of pathology, 248(2), 131-141.
[10] Arai, T., Kuroishi, T., Saito, Y., Kurita, Y., Naruke, T., & Kaneko, M. (1994). Tumor doubling time and prognosis in lung cancer patients: evaluation from chest films and clinical follow-up study. Japanese journal of clinical oncology, 24(4), 199-204.
[11] Werahera, P. N., Glode, L. M., La Rosa, F. G., Lucia, M. S., Crawford, E. D., Easterday, K., ... & Hedlund, T. (2011). Proliferative tumor doubling times of prostatic carcinoma. Prostate cancer, 2011.
[12] Maurizio, M. (2017). Prevalence of malnutrition in patients at first medical oncology visit: the PreMiO study.
[13] Fearon, K., Strasser, F., Anker, S. D., Bosaeus, I., Bruera, E., Fainsinger, R. L., ... & Davis, M. (2011). Definition and classification of cancer cachexia: an international consensus. The lancet oncology, 12(5), 489-495.