(A) Tumor initiation was significantly suppressed by SLC38A2 knockdown (shSLC38A2 #1, #4) compared to control (shGFP) PANC1 cells (2x105) co-injected with hPSC#1 cells (1x106) in a subcutaneous xenograft model. Subcutaneous tumors were monitored bi-weekly by caliper measurement and considered formed if length and width were measured to be ≥1mm each. Each injected flank was considered a possible tumor when calculating percent tumor-free. (B) Tumor initiation was significantly suppressed by SLC38A2 knockdown (shSLC38A2 #1, #4) compared to control (shGFP) PANC1 cells (5x105) in an orthotopic xenograft model. Orthotopic xenografts were monitored by 3-D ultrasound bi-weekly and considered tumors if volume ≥1mm3. (C) Subcutaneous (SubQ) and orthotopic tumor maintenance strategy for deconvoluting tumor initiation versus maintenance effects with acute SLC38A2 deletion. Mice were pre-treated with chow containing doxycycline (625ppm) for three days prior to injection and for five (orthotopic) or seven (subcutaneous) days after injection to allow for SLC38A2 expression during tumor initiation period. Following tumor initiation, mice were switched to new cages with regular chow (no doxycycline). (D) Tumor initiation was rescued by doxycycline-induced SLC38A2 expression in Slc38a2-deficient (sgSlc38a2 #1) HY15549 cells. No significant difference was observed in tumor volume, measured by caliper measurement at day 7 after injection, between control (sgTom) and Slc38a2-deficient (sgSlc38a2 #1) tumors. (E) Tumor volume of subcutaneous control (sgTom) or Slc38a2-deficient (sgSlc38a2 #1) tumors following dox withdrawal after seven day initiation period. Significant tumor regression in Slc38a2-deficient tumors was observed following withdrawal of SLC38A2. (F) Final tumor weight of subcutaneous (left panel) and orthotopic (right panel) control (sgTom) or Slc38a2-deficient (sgSlc38a2 #1) tumors. Subcutaneous tumors were resected and weighed at 20 days (sgTom) or 64 days (sgSlc38a2 #1) after dox-withdrawal. Orthotopic tumors were resected and weighed after 21 days dox-withdrawal. Significance determined with log-rank Mantel-Cox test in A-B; two-tailed t-test in D and F; two-way ANOVA using Sidak’s multiple comparisons test in E. * p<0.05, ** p<0.01, *** p<0.001.