It's probably worth trying to answer this in several parts. I don't want to talk about intrinsic charm at all to start with, because it's complicated enough as it is. Instead I wanted to talk about things related that might help to clarify how this could make sense.
Firstly, I think it's helpful to emphasise that mass is a measure of how reluctant something is to move. The more mass, the harder it is to move -- eg, the more energy required to get it moving at a given speed, or the more force to accelerate it a given rate. The connection with energy is particularly helpful here because you can imagine that if the system has some interactions itself, that might change what something that's looking at the system as a whole measures. In other words, the mass of a system as a whole depends on what's going on inside that system.
Starting, say, with atomic nuclei. So, for example, Oxygen contains eight protons and eight neutrons. Its mass is then naively going to be the sum of these, which is (in atomic mass units)
8*1.0073 + 8*1.0087 = 16.13
but we measure the mass of oxygen in the same units to be about 15.999, which is somehow less. How is this? The answer is that interactions between the protons and neutrons serve to lower the effective mass. This is the "binding energy", which can be described using the method in the wiki page below:
https://en.wikipedia.org/wiki/Semi-empirical_mass_formula
For an intuitive picture of what's going on here, the strong force, the interactions between protons and neutrons, "takes away" mass because the particles *want* to be together. On the other hand, the repulsive force between the protons, the "Coulomb interaction" in that wiki pink, acts to push things apart and increases the mass, by increasing the energy required to hold this system together. I don't see the point in explaining the other two terms, the "asymmetry term" and "pairing term", but they're trying to measure other effects and again can either make it easier or harder to hold the system together. In any case, they alter the internal energy of the system.
The upshot is that the mass of a nucleus is less than the mass of the particles inside it, because they are interacting with each other, and changing the energy of the state. It doesn't really matter how those interactions work, the point is that the mass of a combination of particles is equal to the sum of the individual masses, *plus* or *minus* some corrections.
Linking back to intrinsic charm, with no further details we could at least imagine that the mass of a proton receives a contribution from the mass of the charm quark, but either:
i) *minus* some corrections, so that even if on its own the charm mass is more than the proton mass, those corrections balance it out, or;
ii) have the charm mass multiplied by something that tells us "how much" of the proton is really charm.
I'll try to explore both later, I'd need to study the paper in more detail, but the real point I want to emphasise, again, is that the mass of any combination of things isn't just the sum of its individual pieces.