Psychoactive substances not only can alter neural cells’ function, but may also cause oxidative stress and induce epigenetic modifications, that should be considered as adaptive responses [2,24]. We performed this systematic review to evaluate whether current knowledge about the epigenetic changes caused by drugs in human tissues is enough to allow a translation into forensic practice. Indeed, forensic applications of these kinds of markers could be many, for instance, to evaluate the fitness to drive in the driving license regranting process or to evaluate the history of abuse of psychoactive substances in criminal proceedings [37,38,39,40]. Epigenetic analysis is relatively complex and expensive, but hypothetically it could serve as confirmation test in ambiguous situations or when traditional toxicological methods are limited. For instance, our review showed that epigenetic changes are mainly consistent after long exposure to a specific drug, and in these cases the current gold standard is performing hair analysis. However, this analysis is limited, for example, when the person has short/no hair, when the exposure to the drug is relatively recent, or in the case of extreme hair treatments (i.e., blanching) [41]. So, in these cases, while traditional tests (urine or blood) can only detect recent drug use, in the absence of keratinic matrix, DNA methylation signatures in peripheral blood cells can show signs of past exposure to drugs for extended periods, potentially weeks to months after cessation [42]. Moreover, its sensitiveness is limited for very common substances of abuse like alcohol and cannabis. Other potential forensic applications of drug-induced epigenetic changes could be found in postmortem toxicology, and transgenerational and environmental investigations, for instance. When biological matrices are compromised by decomposition, epigenetic markers could offer an innovative way to detect signs of intoxication. After death, the relative stability of DNA methylation patterns in comparison to drug metabolites makes these markers potentially useful in cases involving long-term substance abuse history [43]. Another potentially interesting use of these epigenetic changes could be projected in every setting where transgenerational or environmental exposure to a substance is suspected. As seen, these changes can be transmitted to subsequent generations leading to increased susceptibility in offspring who were never directly exposed, opening new perspectives about parental liability. In order to highlight the potential forensic relevance of epigenetic biomarkers, we have added a comparative summary (Table 1) outlining their main advantages and limitations in relation to conventional toxicological methods routinely applied in forensic practice. Moreover, where habitat-specific epigenetic signatures are found in specific populations, these findings can be used to reveal exposure to particular chemical pollutants [44]. As shown, current epigenetic information is mainly obtained from animal models and is mainly targeted for clinical purposes. For instance, drug-induced epigenetic changes are being considered for therapeutic purposes: for instance, methamphetamine is able to increase the expression of CCR5 (a chemokine receptor) in myeloid cells and HIV load and, at the same time, inhibits HIV-1 replication in CD4+ T Cells through anti-HIV-1 miRNA [41,45]. Moreover, an interesting use of epigenetic information is also represented by the so-called “epigenetic clock” (i.e., use of epigenetic information as a marker of biological age): Huang et al. reported that illicit drugs (in particular, ketamine) may accelerate this clock in prefrontal cortex and NAc (involved in the drugs’ reward circuitry) [46].

We found that despite there being some forensic literature on applications like age estimation, the actual use of epigenetic analysis for practical forensic toxicology purposes has never been reported [47]. However, as shown by our review, a tentative set of markers could be hypothesized based on the most robust studies (see Table 2).

Non-codifying RNAs (in particular, miRNAs) should be considered particularly promising, because the human brain contains a very high variety of these molecules, since they concur in regulating its development and plasticity [19]. Our review showed that they could be of particular interest to investigate cases of intoxication by alcohol, cocaine, methamphetamines, or opioids, possibly information with sets of CpGs as markers [19,35].

However, many limitations can be already identified. First, other substances frequently influence the epigenetic response to substances of abuse. For instance, in animal models, nicotine was shown to prime the response to cocaine inhibiting histone deacetylase (and thus causing the activation of FosB) in the striatum [48]. Moreover, the effects of substances of abuse like alcohol, cocaine, opiates, and methamphetamine can be also mediated by the concurrent use of antibiotics like ceftriaxone (a third-generation broad-spectrum beta-lactam antibiotic) or butyric acid [49]. Indeed, ceftriaxone increases the transcription of GLT1, thus enhancing the reuptake of the glutamate (an excitatory neurotransmitter) and reducing the drug reward [49]. On the other hand, butyric acid exposure is related to diet and to the molecules produced by aerobic microbial fermentation of polysaccharides in the gut, and may inhibit HDAC, thus reducing the drug reward [49].

Post-mortem use of these markers has its own specific limitations. For example, the tissues used for the analysis cannot be randomly chosen because, even in non-stimulated subjects, different anatomical regions (e.g., NAc and prefrontal cortex) can show different methylation signatures, and even if/when exposed to drugs, the responses tend be more similar [46].

A key limitation is likely to be represented by post-mortem stability of the changes: as shown by our review, epigenetic alterations are usually temporary, and data about the relationship between these changes and post-mortem interval are extremely scarce, although current evidence supports this, at least in the first day post-mortem [36]. Moreover, in many sudden deaths in the young, when attributed to toxic origin, the toxicological history is unknown [50]. This is extremely relevant because, as shown by our review, there can be a difference in the epigenetic changes depending on the pattern of use. For example, in animal models, acute cocaine exposure causes temporary acetylation and phospho-acetylation of histone H4, while repeated cocaine exposure induces persistent H3 acetylation and H3K9 demethylation [51]. Moreover, our review showed that the method of use of the substance is currently underexplored: only one of the reviewed papers was focused on a specific method of use (intravenous injection) [32].

Finally, a significant variable of forensic interest is represented by age. First, because age—as said—influences the epigenome [3,24]. However, age is also relevant to discriminate whether an epigenetic variation is actually “acquired”: epigenetic changes are being considered as one of the explanations of the high heritability of some addictions—like alcohol use disorder—and, for example, epigenetic variations of 65 loci at birth as associated with higher risk of substance abuse during adolescence [52,53,54].