Recently, Azzi et al1 suggested TT virus (TTV) occurs in first-generation recombinant factor VIII concentrate (rFVIII) products with the stabilizer, human serum albumin (HSA) as its source. As these results differ from those of earlier studies2,3 and Baxter's internal findings, we reinvestigated 11 lots of rFVIII (Recombinate), and 13 lots of HSA. Lyophilized rFVIII was reconstituted with distilled water, or HSA used directly. One part of each sample was extracted directly and the other spiked with approximately 300 genome equivalents of TTV (derived from a TTV-positive plasma sample) before extraction, to control for inhibition of the polymerase chain reaction (PCR).
DNA was extracted from 200-μL samples (Blood Kit; Qiagen, Hilden, Germany) and the solution in the quantity described (6 μL)1 subjected to either nested PCR (primer sets untranslated region [UTR] A and B as described1,4), or single-stage PCR (primers T801 and T935, used previously for Baxter's internal investigations5), using a thermally activated DNA polymerase (HotStarTaq, Qiagen). All primer sets are specific for the non–coding region (NCR, also UTR) of the TTV genome.
Both the single-stage and nested PCR reactions were done in 50 μL containing 1 U polymerase, 200 μM dNTP, and 50 pmol each of forward and reverse primer. The samples were overlaid with mineral oil, incubated for 14 minutes at 94°C and amplified for 45 cycles (single-stage PCR) or 35 cycles followed by 25 cycles (nested PCR) in a TRIO-Thermoblock (Biometra, Göttingen, Germany) with the following cycle profile: 30 seconds at 94°C, 30 seconds at 55°C, 60 seconds at 72°C with a final elongation at 72°C for 1 minute. The PCR products were analyzed using a 3.5% low-melting agarose gel stained with ethidium bromide.
Amplification of a positive control sample (DNA extracted from human TTV-positive plasma) with primer pair T801/T935 resulted in a PCR product of 199 bp as expected, with the UTR A primers of 143 bp, and the UTR B primers of 141 bp.
All rFVIII samples were negative with each primer set evaluated. To control for false-negative results caused by inhibition of PCR by the sample matrix, all samples were extracted again and PCR-tested after spiking with a TTV control. All these samples showed positive PCR signals, verifying the results from the unspiked rFVIII samples. Positive samples (TTV-positive human plasma) and negative controls (buffer) also included with each PCR reaction again verified the experimental set-up used.
Additionally, 13 lots of HSA were tested as described above. All samples were negative, with positive results for the same samples after spiking with a low amount of TTV-positive human plasma.
When TTV was discovered in 19976 and suggested to be associated with posttransfusion hepatitis, naturally patients, regulatory bodies, and the plasma products industry were concerned. Further investigations revealed a low viral load in plasma and high prevalence of TTV in up to 82% of plasma donations, and consequently in many plasma pools.2 More refined products (eg, albumin) were, however, consistently TTV-negative.2,3 These findings confirm longstanding experience, derived from numerous viral validation studies, of the Cohn HSA manufacturing process as one with a high capacity for removing viruses. The explanation for the recent unexpected findings1 thus remains enigmatic. Using the same 2 TTV primer sets and a third independent in-house primer system, we confirmed earlier reports2,3 from research groups at regulatory bodies who unanimously found HSA to be TTV-negative.
First-generation recombinant factor VIII concentrates are free from viral contaminations?
Kreil et al (from Baxter BioScience) failed to detect TT virus (TTV) DNA in 11 lots of a Baxter first-generation recombinant factor VIII concentrate (rFVIII) (Recombinate) as well as in 13 lots of human serum albumin (HSA), a difference from our previously reported results.1-1 As regards the rFVIII, the results of Kreil et al are not significantly different from ours (3 out of 13 were positive for TTV DNA). However, with regard to the contamination by TTV in HSA lots, the difference between the results obtained by the 2 groups is more evident. Kreil et al cite, as we too have done, the paper of Pisani et al,1-2 who failed to detect TTV DNA in HSA, in order to strengthen their own conclusions. As we already stressed, Pisani et al used only the N22 polymerase chain reaction (PCR), which is unable to detect a high number of TTV variants.1-3 In this case the difference in the methods used may well justify the different results. In addition, it is well known that, even using the same methods, the results obtained from different laboratories are not fully comparable if international standards are not available and used, as in the case of TTV so far. Furthermore, when the size of the study is so small, a negative result can be affected by a type II statistical error. According to the “rule of 3,” the one-sided 95% confidence intervals of the Kreil et al study are 0 and 27.3.1-4
Similar viral safety problems emerged concerning the possible contamination of several blood products by parvovirus B19. As regards the possibility of B19 contamination, we failed to detect B19 DNA in either rFVII or HSA, even using very sensitive nested PCR, whereas other groups reported different results.1-5,1-6
It is likely that the residual amount of virus (or better, of viral genome) (TTV or B19 virus or perhaps other viruses) in such products after the manufacturing process is very low, near to the limit of sensitivity of the analytical methods available at present. Such a condition, in addition to the lack of standardization of methods, makes a comparison of results obtained in different laboratories very hazardous.
We would like to emphasize again the need for continuous drug-surveillance with prospective protocols of informative hemophiliacs, treated for the first time even with rDNA-derived clotting factor concentrates. In addition, there is also the need to implement the standardization of the molecular methods for the detection of viral contaminations by the development and use of calibrated reference samples.