This study compared two different methods to identify serotype 6C S. pneumoniae; a phenotypic method, the Neufeld test, based on the use of a novel polyclonal factor antiserum 6 d specific for serotype 6C and a genotypic method based on PCR identification of the wci N gene size difference between serotypes 6C and 6A. The two methods provided the same results when testing 96 original serotype 6A isolates.
In addition to the novel antiserum 6 d, antiserum 6b that showed specificity to both serotypes 6B and 6C was altered to antiserum 6b* specific only to serotype 6B and not to serotype 6C. These novel antisera together with the known sera, should provide a sufficient panel of antisera to be able to distinguish the currently known serotypes of serogroup 6. All antisera are commercially available and antiserum 6b obtained from the supplier after 1 January 2009 is antiserum 6b*.
With this study we have shown that it was possible to produce useful polyclonal antisera to be able to distinguish serotype 6C from the other serotypes within serogroup 6. The specificity of the factor serum 6 d was in agreement with a recent validation study of this factor serum performed by a different research group . However, in that study, factor serum 6b and not 6b* was used, which is why they observed serotype 6C to be positive in both factor sera 6 d and 6b.
Very recently, a research group at CDC also showed that it was possible to produce a specific polyclonal antiserum to distinguish serotype 6C from other serogroup 6 isolates following the same strategy as described here; their antisera are, however, currently not commercially available .
One different phenotypic method based on monoclonal antiserum identifying one specific epitope for the identification of serotype 6C has been reported . In contrast, several genetically based methods based on the identification of the wci N gene or sizes of specific PCR fragments have been reported before [2, 7, 10, 6].
With the discovery of serotype 6C, it became clear that new technology, especially genetically based methods, will change the number of known serotypes and the division of groups into serotypes. It also became clear that it is possible to use different methods for determining a serotype. More methods provide possibilities for more laboratories to perform serotype determination of isolates, which is of high value when estimating and evaluating vaccine efficacies and vaccine failures. Different methods for serotype determination will complement each other, but all methods should be compared to one international standardised reference method.
The international reference method and the WHO recommended method for the serotyping of S. pneumoniae is the Neufeld test and the use of type specific polyclonal antisera. With the production of the novel antisera reported here, the Neufeld test can continue to be the international reference method.
In 2007 and 2008, serotype 6C constituted 28.1% (27 of 96) of all original serotype 6A isolates received in the Danish national reference laboratory. Among all invasive isolates from the same time period, serotype 6C constituted 29.6% (24 of 81) of the original serotype 6A isolates. This corresponds to 1.2% of all invasive S. pneumoniae isolates and an incidence of 0.22 per 100,000 persons. This proportion is at the same level as observed in other studies. Among isolates in Cleveland, the U.S., from 1999 to 2007, 6C was observed to constitute 26% of the previous serotype 6A isolates , among invasive isolates from Brazil 1996 to 2007, 34% of previous serotype 6A isolates were found to be serotype 6C , and among Australian invasive isolates previously defined as serotype 6A, serotype 6C constituted 31% . In contrast, serotype 6C was found to constitute only 5% of invasive isolates from children in South Africa in 2005-2006  and only 0.3% among previously defined 6A carriage isolates from Dutch children in 2002 . These studies showed that serotype 6C is widely spread in the world and that it may constitute a higher or lower proportion of the previously defined serotype 6A isolates. The proportion of serotype 6C compared to the other serotypes of serogroup 6 as well as the proportion of serotype 6C compared to all serotypes is expected to change with time and may change differently in different areas with use of vaccines and antimicrobial agents as factors that can provoke serotype distribution changes. Such changes within serogroup 6 have already been observed by the Active Bacterial Core Surveillance Team in the U.S. They tested 16% invasive isolates of original serotype 6A in 1999 to be serotype 6C; however, this changed to 65% and 69% in 2006 and 2007, respectively. The change corresponded to a change in the incidence of 6C invasive pneumococcal disease from 0.22 in 1999 to 0.58 in 2007 . The dramatic change was related to the introduction of the 7-valent pneumococcal vaccine in 2000 causing a decline in the incidence of both serotype 6B and 6A cases, but an increase in the proportion of serotype 6C cases [ref ]. This was probably observed, as the vaccine protects against invasive infections with serotype 6B, but in addition the vaccine has been shown to have a high level of cross-protection against serotype 6A but not 6C . Similar changes in the distribution of serotypes within serogroup 6 can be expected in Denmark in the future, as the vaccine was introduced in 2007.
The sample origin (sample material, age and sex of person) and antibiotic resistance level of serotype 6C were quite similar to that found for the other serogroup 6 isolates. The proportion of antibiotic non-susceptible 6C was low compared to observations done in the U.S. [5, 6, 17], but that may also change as observed by the Active Bacterial Core Surveillance Team in the U.S. after introduction of the vaccine [ref ].
Among all Danish serotype 6C isolates from 2007 and 2008, ten different STs belonging to nine CCs were found and among the three old 6C isolates three different STs and CCs were found. The serotype 6C isolates in Denmark were, therefore, regarded as diverse. Except for ST386 and STnew1, all of the serotype 6C STs had been identified before in relation to serotype 6A from a number of different countries (Table 3). It would be interesting to re-analyse those serotype 6A isolates as they may turn out to be serotype 6C with the current knowledge. Thus ST481 and CC460 had only been related to serotype 6A, they may, therefore, be true serotype 6A and not 6C if re-analysed. Four of the STs were previously found in relation to serotype 6C, all from geographically distant countries like Portugal, Australia and the U.S. (Table 3). One ST, ST386 had previously only been found in relation to serotype 6B in several countries (Table 3). All of the CCs, except for CC386 and CC3304, had also been observed in a number of countries in relation to serotypes 6A or 6C (Table 3). CC386 had only been found in serotype 6B isolates and in addition to 6A and 6C, CC473 had also been observed in serotype 6B. CC395 had in addition to serotypes 6A and 6C also been observed in serotype 23F whereas CC3304 was previously only observed in serotype 4 (Table 3). All of these observations confirm the suggestion by Carvalho et al. 2009 that serotype 6C strains arose from independent recombination events. Thus, in contrast to that study suggesting these involved parental strains of serotypes 6A and 6C, this study suggests that these events involve at least serotypes 6A, 6B and possibly also serotypes 23F and 4 as parental strains. With this study, it was also confirmed that serotype 6C is an "old" serotype being present among S. pneumoniae isolates in Denmark for at least 48 years. This time span would allow the possibility that 6C strains may have arisen from several different independent recombination events involving different parental strains.