Dittrichia viscosa uptake and translocation of the metalloid As is not fully understood and some data are contradictory, but its adaptability to this pollutant is known and is dependent on its genetic variability. D. viscosa is not a hyperaccumulator plant, but it can grow in high-drought conditions while still producing large biomass, even tolerating significant concentrations of As3+ and As5+. In spite of these remarkable characteristics, adaptive modification of performances is not predictable in wild populations. In previous work, we established experimental clonal populations to perform a functional study on the aquaporin NIP1.1. Here, we propose a strategy to select a clonal population of D. viscosa with a defined phenotype related to As tolerance and to reduced NIP1.1 expression levels for phytoremediation applications. From the previous work, we selected four independent clones, two of them belonging to the weak population (W8 and W9) and the other two belonging to the strong population (S1 and S3). The weak and strong populations differ for a different expression ratio root/shoot of DvNip1;1 that brings a different tolerance to As presence. The stress response of the populations, revealed by the CAT enzymatic test, was statistically correlated to the clones, but not to As uptake. Performance of the selected plants on a second unrelated metallic pollutant, Cd, was evaluated, showing that Cd uptake is also independent from the tolerant phenotype. In vitro culture methods using solid media and temporary immersion bioreactors were compared to propose an optimized combined protocol. The procedure yielded propagation of genetically stable tolerant clonal lines with good uptake of As and Cd. The plants, mass-produced with the developed in vitro protocol, were able to maintain their acquired abilities and are potentially able be later applied in phytoremediation or contaminated areas’ re-naturalization.