This is a very peculiar case. If we apply the CPCA rules, this substance seems a very potent nitrosamine:
Count of hydrogen atoms on each α-carbon [2, 2] and corresponding α-Hydrogen Score is 1
Chains of ≥5 consecutive non-hydrogen atoms (cyclic or acyclic) on both side,Individual Deactivating Feature Score +1
Aryl group bonded to α-carbon, Activating score -1
Score: 1
Potency Category 1 : AI = 18 ng/day
On the other hand, the substance was found negative in an old carcinogenicity study on rats (Druckery et al., 1967):
N-Nitrosodibenzylamine (Lhasa Carcinogenicity Database)
A summay of the corrent knowledge has been reported in the
EFSA - Risk assessment of N-nitrosamines in food
NDBzA is reported as unable to induce tumours by Druckery et al. (1967), that also reported problems with the solubility of the compound.
However, all genotoxicity studies and expected metabolism would predict carcinogenicity.
NDBzA is genotoxic, being positive in the Ames test and in several other assays, including the MutaMouse assay. In the latter system, the predominant type of NDBzA-induced mutations were transversions, mainly GC > TA. These may arise from unidentified DNA adducts with benzylation possibly being the primary mechanism. Interestingly, in the Ames test, NDBzA is positive not only in the usual TA100 and TA1535 strains (as the other N-NAs) (small adducts, point mutations), but also in strain TA98 (large adducts, frameshift mutations).
No metabolic data are available; however, given the above MutaMouse results and in analogy with results reported by Moschel et al. (1979), the ability to benzylate DNA can be hypothesised.
Regarding the estimation of TD50 by read across, a partially similar structure is the carcinogenic N-methyl-N-nitrosobenzamide (CAS 63412-06-6), with TD50rat = 3.23, for which a related type of DNA damage, DNA benzoylation, could be hypothesised. This agrees with the assignment to NDBzA of a
level of concern of LOW-MODERATE carcinogenic risk by Oncologic
If we consider the EFSA position, the TD50(rat) of 3,23 mg/kg/day will lead to a lifetime AI of 3.23 µg/day (3230 ng/day).
I don’t think that this read-across with N-methyl-N-nitrosobenzamide will be easily accepted by the authorities; on the other hand, the CPCA AI of 18 ng/day is very conservative.
Probably the truth is somewhere in the middle, but it is very hard decide where…