Not common, but not all that rare. It's the result of what's called a "bud sport"--essentially a mutation of a cell in the growing tip that eventually grew into that flower. When that happens in woody plants, plant breeders take cuttings or grafts from the affected stem, to get new colors or forms of flowers or foliage. 'Chicago Peace' rose is a famous example of the uses of the phenomenon. In this case, if you wanted to save the new color, you would have to save seeds from that flower, and hope that some of the seedlings carried the new gene.

Bud sport indeed. The cell that had changed (genetic or epigenetic change) and led to the sport is at the base of the petals and the chances of the same change being in the ovary/ovum is very small as the actual flowers are the tiny units in the centre. In any case the progeny would be either peach-orange or pink (assuming self-pollination) as the mosaic-effect would not be preserved into the next generation.

This is because each fertilised seed is a single cell and such a flower requires two populations of differing cells (called a mosaic). Mosaic plants are always propagated vegetatively.

Thanks, Buddleja! I should have emphasized that the multicolor effect would not be preserved. I was also assuming that since a whole section of the inflorescence was involved, that a segment of the flowering stem was also, including a segment of the fertile flowers in the center. Even if that is so, the seedlings might not come out the color that you want, since part depends on what flower the stigma was pollinated by. The true color might not show up until the second generation.