Dairy ewes are less prone than cows to milk fat depression (MFD) but suffer this syndrome when marine lipids are added to their diet to improve milk fatty acid (FA) composition. This is very detrimental as most ovine milk is used for cheese manufacture. However, there are large individual differences in MFD severity; the reasons behind this variability being uncertain. This study was conducted in lactating sheep to test the hypothesis that differences in milk concentration of antilipogenic FA or in the transcriptional regulation of mammary lipogenesis may account for that individual variation. We used 15 ewes receiving, for 35 d, a total mixed ration supplemented with 0 (control; n = 5) or 20 g of fish-oil/kg DM [10 animals were selected out of 22 and divided in those showing a strong (RESPON+, n = 5) or slight (RESPON-, n = 5) MFD]. Milk production and composition, and milk FA profile were recorded for 3 consecutive days before and after treatments. Candidate gene expression was analyzed by quantitative reverse transcription-PCR on mRNA isolated from milk somatic cells collected before (d −2 or −1) and after (d 34 or 35) the dietary treatments. Data were analyzed with the MIXED procedure of SAS 9.4 using orthogonal contrasts. Milk production was not affected by the diets (P > 0.10) but milk fat concentration decreased by 25.4% in RESPON+ and 7.6% in RESPON- (P < 0.001). Supplementation with fish oil enhanced (P < 0.01) the milk content of both potentially healthy FA (e.g., cis-9,trans-11 CLA, trans-11 18:1, or very long chain n-3 FA) and antilipogenic FA (e.g., cis-9 16:1, trans-10 and cis-11 18:1, trans-10,cis-15 18:2, trans-9,cis-11 and trans-10,cis-12 CLA, and 10-oxo-18:0). Nevertheless, differences between RESPON- and RESPON+ were hardly detected. Consumption of the MFD-inducing diet was accompanied by reductions in the mRNA abundance of ACSS2, FASN, LPIN1 and INSIG1 (P < 0.10), but only SCD and GPAT4 tended to differ between RESPON+ and RESPON- (P < 0.10). These results oblige to reject the hypothesis and conduct a thorough evaluation of individual responses at systemic, ruminal, and mammary levels to explain the individual variation in MFD severity.