Tuberculosis (TB) is the second cause of death from a single infectious agent, the M. tuberculosis bacillus. Nearly two billion people are infected and about 8.7 million new cases and 1.4 million deaths were reported by the World Health Organization (WHO) in 2013. Despite the availability of effective treatment, the alarming emergence of multidrug resistant (MDR) strains (with 310.000 estimated cases in 2011 among notified patients with pulmonary TB), simultaneously resistant to the two most effective anti-TB drugs, isoniazid (INH) and rifampicin, has urged the need to develop new molecular scaffolds, either structurally original or based on old and active drugs. The aim of this review is to summarize the current status of different QSAR based strategies for the design of novel anti-TB drugs based upon the most active anti-TB agent known, INH. A case study puts in evidence that the judicious application of quantitative structure- activity relationships can be successfully used to rationally design new INH-based derivatives, active against INH-resistant strains harboring mutations in the most frequent resistance related target (katG), and therefore develop candidate-compounds against MDR-TB, thus revisiting the unique effectiveness of INH against TB.