Under the usual definitions, there is absolutely no possibility that 7 is a prime number, in this or any other universe.
The usual definitions are, say (among many equivalent ways of stating them), that
- a positive integer n is prime if n ≠ 1 and n has no positive divisors other than 1 and n, and
- an integer d is a divisor of an integer n if there exists some other integer c such that n = cd.
- (Plus some more definitions, in more detail, of "integers", of products, etc.)
So the fact that 7 is a prime number is simply stating that none of the numbers 2, 3, 4, 5, or 6 is a divisor of 7, which is true.
The existence of other possible worlds has no bearing on whether 7 is a prime number, because the "world" that is of interest to mathematics is completely defined by the mathematical axioms. (A question of philosophical interest is what relation this mathematical world has to our real world -- in another universe this mathematical world of integers may be less relevant to that universe -- but that is a separate question entirely, unrelated to whether 7 is a prime number or not.)
Note that even in the case of the square root of -1, there was no previously beleived assertion that was discovered to be false by the introduction of complex numbers: -1 is still not the square of any integer, nor of any rational number, nor of any real number, and will never be, and we can prove all of these with the usual axioms. It so happens that there exists an extension of the set of real numbers in which it happens to make sense to speak of a "square root" of -1: but to do so we must also extend the meaning of "square root" (the domain and range of this function).
Returning to the case of 7, there is a very simple alternative way to get a mathematical "world" in which 7 is not prime: simply move from the ring of integers to the ring of rational numbers. Here, it is possible to extend the definition of prime numbers (everything carries over) to show that 7 is not prime, because for instance it is the product of two numbers (3/2) and (14/3), to pick just one example. (In fact, in the ring of rational numbers, we can show there are no prime numbers at all, as every p/q can be written as the product of 2p/q and 1/2, say.)
As a less trivial example, consider the ring of numbers (a + b√-3), where a and b are both integers. The good old integers are a subset of these, got by taking b = 0. Here, though it may not be obvious at first, it so happens that 7 is not prime, because 7 = (2 + √-3)(2 - √-3). However, there are other prime numbers in this ring: you can prove that 5 is a prime number even in this bigger ring.
In either of these examples, nothing has changed about the properties of 7. Only the definition of "prime number" has changed, and it should not be surprising that with a different definition, things can change. However, in any parallel universe, in the ring of integers and under the usual definitions, 7 will continue to be a prime.