Consensus guidance divided into points of general application and those more relevant to clinical trials, showing strength of recommendation (A–D) based on available evidence, according to the scale (A–D) recommended by the Oxford Centre for Evidence-based Medicine23
| Point | Strength of recommendation | Number of respondents | Mean score (SD) | % ≥6 | |
|---|---|---|---|---|---|
| General guidance | |||||
| 1 | The minimum number of minor salivary glands is suggested to be four (six if small), and should be surgically separated | D | 39 | 8.0 (2.4) | 82 |
| 2 | The minimum surface area of gland sections examined should be 8 mm2 | D | 39 | 7.5 (1.9) | 82 |
| 3 | If the first cutting level is inconclusive, or in the context of a clinical trial, consideration should be given to including two additional cutting levels at 200 µm intervals (typical focus diameter is <50 μm) in order to increase the surface area | C/D | 37 | 8.2 (2.0) | 92 |
| 4 | Care should be given to preparation of paraffin blocks, with smaller glands set higher to allow midspecimen sampling during cutting | D | 38 | 7.5 (2.1) | 87 |
| 5 | Histological examination should determine whether there is FLS present. Attribution of FLS, or possible FLS, should be followed by calculation of a focus score | B | 39 | 8.8 (1.4) | 95 |
| 6 | The extent (absent, mild, moderate, severe) of atrophy, fibrosis, duct dilatation and non-specific chronic sialadenitis, in addition to the presence or absence of FLS, should be reported | C | 39 | 8.5 (1.7) | 92 |
| 7 | Calculation of the focus score should include the whole of the glandular surface area in the denominator, to avoid introduction of bias | D | 39 | 8.3 (1.6) | 95 |
| 8 | The presence or absence of germinal centre-like structures and lymphoepithelial lesions should be reported | C | 39 | 9.5 (1.0) | 97 |
| Guidance relevant to clinical trials | |||||
| 9 | The Focus score should be recorded, and the area of individual foci should also be summed and divided by glandular area to give a more quantitative indication of the extent of glandular infiltration | C | 38 | 7.5 (2.5) | 76 |
| 10 | Once FLS has been confirmed, all foci should be included in the Focus score and in area of foci calculations, even when adjacent to abnormal acini or ducts, to avoid introduction of bias | D | 38 | 7.3 (2.6) | 76 |
| 11 | Staining for CD3, CD20 and CD21 should be included, and the presence of germinal centre-like structures should be reported as the proportion of foci with both T/B-cell segregation and follicular dendritic cell networks. Consideration should be given to reporting the mean B/T cell ratio in foci | C/D | 38 | 8.1 (2.0) | 89 |
| 12 | Scoring should be undertaken by two trained observers who have reviewed a reference slide set, and with reporting of intraobserver and interobserver variability | D | 38 | 8.9 (1.9) | 95 |
| 13 | Samples should be scored blind to subject and order | D | 36 | 8.8 (2.1) | 94 |
| 14 | High-resolution image morphometry of each sample should be stored | D | 38 | 8.2 (2.0) | 89 |
| 15 | Given the stable or slowly progressive nature of the histological features, baseline biopsies may be substituted with prior biopsies to reduce the number of biopsies required. However, given the limited evidence available, these should have been acquired no longer than 1 year prior to baseline | C | 38 | 7.8 (2.0) | 87 |
| 16 | The optimal period of time for rebiopsy has not been established and will depend on the agent employed. | D | 39 | 8.3 (1.6) | 92 |
The level of agreement (0–10 scale) among participants is also shown, represented by mean scores and the percentage of respondents who scored the point ≥6/10.
FLS, focal lymphocytic sialadenitis.