I think you're asking at least two different questions.
(i) what meta-analyses of treatment for "childhood trauma" have been done, and what have they shown
(ii) what evidence is there that a single supportive relationship will be sufficiently therapeutic in order for one to not need other forms of treatment
I think both of these are really great questions BUT one would need to define ones terms very clearly beforehand -- for example, childhood trauma is mentioned as if it was a single thing -- but it isn't, and herein lies a real problem -- what is "childhood trauma" -- not all people go through what has been called in the past a Criterion A Trauma, using DSM criteria for PTSD; and there have been changes to what is called PTSD -- and there will be further changes with the introduction next year as currently predicted, but it's been delayed before) of ICD 11, and then there will be different disorders, at least under ICD 11, called PTSD, and Complex PTSD. Yet people may, or may not, suffer either of these, depending on various combinations of genetic, physiological, and environmental/contextual factors.
But why stop with considerations of formal criteria for just these two disorders, when the research shows that sub-threshold stressors can result in lasting impairments of a range of different negative outcomes, including brain impairments. And therefore of what does "treatment" comprise -- treatments for fear responses (re-experiencing, avoidance, hypervigilance, etc), treatments for emotion and self dysregulation; treatments, treatments for relationship difficulties (relationship with self, and with others), etc etc. For which of these does a supportive relationship suffice to prevent problems -- unlikely to be effective with all problems.
Also, don't forget that something that is absolutely crucial for successful outcomes is the quality of interpersonal / professional relationships between the "client" and their principal and other "therapists", in relation to outcomes, and IMHO that is something that has not been adequately explored yet. I can give you some guidelines about this if you like, but there have been some really good contributions by others on this "list" -- see e.g., recent contributions by Marcia Hall.
But thanks for the question -- it'll make us think, and not just search the literature -- which I don't believe has reached this level of "maturity", yet, but may I bring to everyone's attention a recent interview with Marylene Cloitre, one of the foremost therapy researchers in the area (the actual url is VERY long, so I've used a shortener) https://goo.gl/zVjzXz It goes outside just ACEs but it's very consistent with the "whole of life" perspective in recent blog postings
Incidentally, if you're interested in exploring treatments at a more "molecular" level -- discrete areas of application -- then may I suggest accessing Cochrane Reviews www.cochrane.org/what-is-cochrane-evidence
Best wishes to you
Russell J. Wilson
B.Sc.(Hons), M.App.Psych (Clin) (UQueensland)
Retired Clinical Psychologist