Bengston & Blažek defend the Dene-Caucasian hypothesis against the recent claim by I. Cašule that Burushaski is related to Indo-European. As part of their defense, they suggest (p. 55) etymological links between Burushaski and North Caucasian sibling terms.
Burushaski has a version of a typologically rare sibling type 8 (see Dziebel, The Genius of Kinship, 290). North Caucasian languages have type E, with the same underlying root for both brother and sister, which is two mutational steps down from type 8. The Burushaski term for ‘brother of male’ and ‘sister of female’ fits well with the North Caucasian root underlying the brother and sister terms. Bengston & Blažek see the same root behind one of Basque sibling terms and the only term for ‘sibling’ in Ket. The Dene-Caucasian hypothesis has not been well-received by the mainstream academic establishment, but it’s infinitely more robust than the Casule proposal. In light of the rarity of type 8 globally, it’s noteworthy, as I pointed out in The Genius of Kinship, that Basque, another member of the putative Dene-Caucasian stock, shares sibling type 8 with Burushaski. The reduction from type 8 to type E (North Caucasian) or A (Ket) is consistent with the global phylogeny of sibling sets.
Bengston, John D., and Václav Blažek. 2011. On the Burushaski-Indo-European Hypothesis by I. Cašule. Journal of Language Relationship 6: 25-63.