Impact of mouse model tumor implantation site on acquired resistance to anti-PD-1 immune checkpoint therapy

Abstract

Introduction: The use of tumor subcutaneous (SC) implantations rather than orthotopic sites is likely to induce a significant bias, in particular, in the field of immunotherapy.

Methods: In this study, we developed and characterized MC38 models, implanted subcutaneously and orthotopically, which were either sensitive or rendered resistant to anti-PD1 therapy. We characterized the tumor immune infiltrate by flow cytometry at baseline and after treatment.

Results and discussion: Our results demonstrate several differences between SC and orthotopic models at basal state, which tend to become similar after therapy. These results emphasize the need to take into account tumor implantation sites when performing preclinical studies with immunotherapeutic agents.

Keywords: orthotopic; MC38; anti-PD-1; preclinical model; subcutaneous.

Figure 1

Sensitivity to anti-PD-1 treatment on the MC38 subcutaneous versus orthotopic implantation models. (A) Treatment administration for C57Bl/6 mice SC implanted with a MC38 tumor fragment. The mice were randomized when the tumor volume reached 150 mm3 then treated or not with anti-PD-1 (BioXCell, BE0146, RRID: AB_10949053, 12.5 mg/kg per week, intraperitoneal (i.p.)). (B) Tumor growth in mice SC grafted with MC38, untreated or treated with anti-PD-1. (C) Individual values of the last measurement point of the mice having received treatment or not. (D) Individual curves of tumor growth in mice SC grafted with MC38, untreated or treated with anti-PD-1. (E) Treatment administration for C57Bl/6 mice orthotopically implanted with a tumor fragment of MC38. (F) Representative image of MC38 tumor implanted in the cecum on D4. (G) Individual curves representing the weight change in grams of mice orthotopically grafted with MC38, untreated or treated with anti-PD-1. (H) Individual values of the weight of the intestines at D18 of the mice having received treatment or not. (I) Photographs of untreated and treated intestines grafted with orthotopic MC38 at D18. Data represent mean tumor volume and error bars represent ± SEM. Results are representative of three independent experiments, n = 5 per groups, **p < 0.01, Mann Whitney t-test.

Figure 2

Induction of acquired resistance to anti-PD-1 therapy in the SC versus orthotopically implanted MC38 model. For the SC model implantation, 5.10⁶ tumor cells were injected in each animal. When tumors reached a volume of 150 mm3, mice were randomized and treated with aPD-1 (BioXCell, 12.5 mg/kg per week, i.p.). For the orthotopic model implantation, fragments of tumors were implanted in the cecum. Two days after surgical implantation mice were randomized according to body weight and treated or not with aPD-1 (BioXCell, 12.5 mg/kg per week, i.p.). For both types of implantations, fragments of tumors displaying a primary response to aPD-1 were then implanted into new groups of tumor-naive mice and treated once a week to maintain selection pressure. At least five and seven passages were necessary to induce acquired resistance for SC and orthotopic implantations, respectively. (A) Tumor growth in mice grafted with MC38 subcutaneously, untreated or treated with anti-PD-1. (B) Overall survival of mice grafted with orthotopic MC38, untreated or treated with anti-PD-1. (C) Tumor growth in mice grafted SC with MC38 model that had been rendered resistant to anti-PD-1. (D) Overall survival of mice grafted orthotopically with MC38 model that been rendered resistant to anti-PD-1. Data shown represent mean tumor volumes and error bars represent ± SEM. Results are representative of three independent experiments for the SC model and two independent experiments for the orthotopic model, n = 5 per groups, **p < 0.01, Mann Whitney t-test. ns: non significant.

Figure 3

MC38 sensitive tumor TIME at basal and after anti-PD-1 therapy. Flow cytometry experiments were performed when tumors reached 200 mm³ or two days after implantation for basal stage for SC and orthotopic implantation respectively and four days after second treatment for both. (A–D) Sunburst plots showing the proportion of CD45+ immune infiltration. (A) Basal stage for SC implantation, (B) Basal stage for orthotopic implantation, (C) Anti-PD-1 treated group for SC implantation, (D) Anti-PD-1 treated group for orthotopic implantation. (E, F) Histograms shown mean of percentage values and error bars are SEM of basal versus under treatment for (E) SC model and (F) Orthotopic model. Flow cytometry plots represent a pool of five tumor samples. Significant decreases and increases were assessed by a two-way ANOVA statistical test, with Bonferroni post hoc test. Statistical significances are indicated as follows: *p < 0.05, ***p < 0.001, ****p < 0,0001.

Figure 4

MC38 resistant tumor to anti-PD-1 TIME at basal and under selection pressure of therapy. Flow cytometry experiments were performed when tumors reached 200 mm³ or two days after implantation for basal stage for SC and orthotopic implantation respectively and four days after second treatment for both. (A–D) Sunburst plots showing the proportion of CD45+ immune infiltration. (A) Basal stage for SC implantation, (B) Basal stage for orthotopic implantation, (C) Anti-PD-1 treated group for SC implantation, (D) Anti-PD-1 treated group for orthotopic implantation. (E, F) Histograms shown means of percentage values and error bars are SEM of basal versus under treatment for (E) SC model and (F) Orthotopic model. Flow cytometry plots represent a pool of five to ten tumor samples. Significant decreases and increases were assessed by a two-way ANOVA statistical tests, with Bonferroni post hoc test. Statistical significances are indicated as follows: **p < 0.01, ***p < 0.001, ****p < 0,0001.