I think you may have misinterpreted some of the research results.
Dr. Shelton’s study did say that about 89 percent of the dogs in her study did remit from MG within an average of 6.4 months, and that the remaining 11 percent that did not remit had a neoplasia (a fancy word for a tumor), presumably a thymoma (though the abstract did not explicitly state this, though I think that is a reasonable assumption).
There are several points to make here, all statistical in nature. First off, the 6.4 months is an average, of all the dogs in the study (50 or 60, I seem to recall). The nature of an average is that not all (or even most) of the individuals are close to that. There will generally be a spread of values. It would have been nice if the study abstract had also given the standard deviation, which helps indicate how widely the values are spread about the mean. Typically, for a mean (same thing as average) of anything, there will be a number of individuals clumped around the mean, but rarely exactly on it. But there will also be many other individuals further away from the mean, but progressively fewer, the further away you get from the mean.
I believe I read elsewhere that in Dr. Shelton’s MG study, some dogs took a year or a year and a half to remit from the MG. My own Emmy took 20 months to completely remit. So the fact that a dog has not remitted from MG in some particular period of time is no proof it has a thymoma. If has been a long time, it might be worth checking for a thymoma, but the only proof of a thymoma is finding one. Another point is that Dr. Shelton’s study was only one study, and a small one (only 50 or 60 dogs), at that. If you did the same study a number of times, the results would probably vary slightly, which is the nature of probability. And some might well have some dogs that did not remit but did not have a thymoma either.
A small study cannot accurately detect or measure very small percentages. In a study of 50 dogs, say, you cannot measure percentages under 2 percent. So if 1 percent or half a percent of dogs with MG do not remit (or take a very long time to remit) and yet do not have a thymoma, such a study cannot accurately measure this. Many times, there would be no such dogs, giving a zero percentage, which would be too low. But occasionally one (or even two) such dogs might show up in such a study, giving 2 percent (or 4 percent), which would be too high. The point is, that for small studies (and even many large ones), you have to take the results with a grain of salt. The best you can hope for is that the results are roughly indicative, to within maybe a few percent. If a very large version of Dr. Shelton’s study (with thousands of dogs) were done, we might find that the true percentages might be something like 87.6 percent remit in an average of 7.1 months, 1.3 percent have not remitted within 5 years (but have no neoplasia), and 11.1 percent have a neoplasia and do not remit.
For example, in national political polls and the like, if they bother to say so, they usually note that the results are accurate to within maybe 5 percent. And if they really fully disclose, they say that is at the 95 percent confidence level (or some such). Which means that 95 percent of the time, the results will not be off by more than 5 percent. But 5 percent of the time, they might be off by more than 5 percent. Statistics is nothing if not fuzzy, is almost never exact, so you need to interpret such results carefully. I expect your Nahla was in the category of not remitting, or not remitting in a very long time, but without having a thymoma, but that Dr. Shelton’s study was simply too small to detect the category.