"The results indicate that only two medications, fluoxetine and risperidone, can be considered as promising evidence-based practices for adults with ASD [autism spectrum disorder]."
So said Lauren Taylor  and her (systematic) review of what might work, pharmaceutically speaking, when it comes to managing "behavioural disturbance in adults with ASD." Including over 40 studies examining psychopharmacology in adults diagnosed with ASD, Taylor concluded that many medicines/formulations did not cut the scientific mustard in this area. Further: "The subsequent establishment of clinical guidelines for medication use in this population is essential."
My first thoughts when I came across the Taylor paper was the guidance provided here in Blighty with regards to medication and autism (see here). To quote from the NICE document as it currently stands: "Do not use antipsychotic medication for the management of core symptoms of autism in adults. Do not use antidepressant medication for the routine management of core symptoms of autism in adults." One might argue that the review detailed by Taylor does not just focus on 'core symptoms' but rather a wider recognition of "psychiatric and behavioural disorder in adults with autism" in light of what is known about comorbidity (see here for example). Based on this assumption, there are indeed various levels of evidence to suggest that for some at least, time-sensitive trials of psychopharmacology might be useful where autism is mentioned.
But just before the floodgates are opened to widespread use of preparations such as fluoxetine and risperidone 'for autism', a word or two of caution. First and foremost are a few questions that everyone should be asking when confronted with so-called challenging behaviours (behavioural disturbances?) when it comes to autism: 'Is the behaviour something new?' and 'What might the triggers be?' I say this because I've covered this topic before, and sometimes one needs to turn detective before reaching for the medicines cabinet (see here). Second, the advantages of such medicines use has to be always balanced against potential side-effects from medication. Not to come down to hard on something like risperidone for example, there is a volume of research suggesting a possible link between this medicine use and elevated prolactin levels for example (see here). Monitoring is key. Finally, the idea that there is increased recognition of 'psychiatric comorbidity' being potentially over-represented alongside a diagnosis of autism is an important one for preferential screening (see here). When emphasizing the possibility of a heightened risk of schizophrenia or psychotic disorder when it comes to the autism spectrum (see here), one might look to the wider literature for clues about how one might mitigate such risk (see here and see here for example). I say that last point with the knowledge that within the massive heterogeneity that is autism, not every person is going to transition to schizophrenia or other psychotic disorder and not every experience of this phenomenon when it does happen, will be the same.
And like buses, yet more on the issue of psychopharmacological intervention for autism . Conclusion: "Overall, randomized, placebo-controlled studies of medications for the treatment of ASD are scarce."
 Taylor LJ. Psychopharmacologic intervention for adults with autism spectrum disorder: A systematic literature review. Research in Autism Spectrum Disorders. 2016; 25: 58-75.
 Accordino RE. et al. Psychopharmacological Interventions in Autism Spectrum Disorder. Expert Opin Pharmacother. 2016 Feb 18.
Taylor, L. (2016). Psychopharmacologic intervention for adults with autism spectrum disorder: A systematic literature review Research in Autism Spectrum Disorders, 25, 58-75 DOI: 10.1016/j.rasd.2016.01.011