3. Results/Discussion

3.1. An Alternate Use of TMA Technology
TMA technology combined with digital pathology can help strengthen biomarker research [10]. Recent advances allow precise histological areas to be digitally annotated onto scanned tissue slides. These annotations are then used to identify specific regions from the corresponding tissue block that should be punched out and transferred to the TMA. Another possible application of this technology is to punch out and transfer tissues directly into tubes, which can then be used for subsequent molecular analysis. However, the possibility for cross-contamination between samples has not yet been addressed. This contamination could arise from tiny amounts of material remaining on the punching device that can be transferred from one tissue sample to the next.

3.2. Cross-Contamination Assessed Using Mycobacterium Tuberculosis Assay
To assess cross-contamination between punches, we alternate mycobacterium tuberculosis-positive and negative tissue blocks in an automated tissue arrayer with a mechanical cleaning step in between. Tissue blocks were punched out sequentially. The presence of mycobacterial DNA in the tissue punches was analyzed by PCR using a primer pair specific for IS6110, an element, which is present at multiple copies in the genome of mycobacterium tuberculosis complex [11]. This method is proven to be highly sensitive as it allows for the detection of as little as 1–2 mycobacteria. A positive mycobacterium result would be indicated by a peak at 123 bp in the corresponding electropherogram. A fragment of 123 bp was clearly detected in the DNA from punches from mycobacterium tuberculosis-positive tissues (Figure 3A,C,E). In contrast, mycobacterium-negative tissues punched out in between, were found to be negative (Figure 3B,D,F). Thus, there appears to be no transfer of mycobacterium tuberculosis from the positive to the negative case.

3.3. Cross-Contamination Assessed Using KRAS Analysis by Pyrosequencing
Since the detection of mycobacterium in the negative samples may be influenced by the level of infection, we carried out the analysis of KRAS mutational status on colorectal cancers. Using only a manual punching device without cleaning step and alternating between tissue samples, no cross‑contamination could be seen. When analysis of KRAS gene mutation was performed using the automated device with cleaning step, again no cross-contamination of tissue samples was found. All mutated samples carried the same mutations, while all WT samples were clearly WT (Figure 4).

3.4. Impact of the Finding
Researchers working with modern TMA technologies have a growing interest in the possible application of TMA instrumentation for purposes other than TMA block construction. Especially, a combined digital pathology/TMA approach is ideal for assessing issues of heterogeneity and allows very precisely annotated histological regions to be transferred either to TMAs or to tubes for storage and/or analysis of nucleic acids. With or without a digital pathology platform, neighboring punches side-by-side for TMA construction and simultaneous molecular analysis is clearly advantageous in the era of molecular pathology.
This study was motivated by the growing concerns over cross-contamination between samples and appears to be the first study to date to address this issue. Using two different PCR methods and alternating between positive/mutated and negative/wild-type samples, our findings underline no evidence of transfer of material from one tissue block to the next.
Figure 3  Electropherograms used to evaluate the presence of mycobacterium in donor blocks from a previously confirmed positive and negative tissue sample. (A) positive control and (B) negative control. Tissue samples were alternated in the tissue arrayer three times leading to the following results: (C,E,G) mycobacterium tuberculosis samples and (D,F,H) mycobacterium tuberculosis negative samples. No evidence for cross-contamination between samples.

3.5. Conclusion
To conclude, our finding suggests that TMA instrumentation is appropriate for use as an accessory to molecular applications. No cross-contamination appears to occur between samples punched with the same device, although a cleaning step in between donor blocks is still recommended.
Figure 4  Pyrograms highlighting results from an analysis of KRAS exon 2 codon 12 and 13. (A) G12D mutation; (B) WT; (C) G12V mutation; (D) WT; (E) G13D mutation; (F) WT.