I would like to establish a immunotherapy model by co-culturing human tumor cell lines with HLA-matched T cells. In particular I am interested in non small cell lung cancer (NSCLC) cell line Calu-1.The HLA phenotype for Calu-1 cells is as follows: HLA A10, A11, B15, Bw35. I need at least partially matched HLA-phenotype, i.e. matching the A-subclass.
Is there any company that provide PBMCs which are characterized for HLA phenotype?
Thanks,
Alessia
Hi Alessia. I don't know of a company that does this, and it would be quite difficult given the HLA disparity in the population to find blood products of a specific HLA type. In the world of clinical stem cell transplantation, it takes millions of people in a reference donor pool in order to find an HLA-matched unrelated donor for any given patient, and even so a large number of patients do not have a donor. One option (though not using Calu-1 cells) might be to try and establish a tumor cell line from a patient for whom you also have blood samples, however this may not meet your experimental need.
If you discover a group that will produce HLA-matched blood product samples for research, however, I would buy from them as well.
-Benjamin
Hi Benjamin,
thank you for your answer. Because I don't need to transplant the cells but only to perform an in vitro experiment (co-culture of primary tumor cells with PBMC), for me also a partial match it would be ok. Calu-1 are from a male Caucasian patient age 47 with grade III, epidermoid carcinoma (non small cancer cell line). The HLA phenotype is A10, A11, B15, Bw35 . I think that also only matching the A-subclass should be ok. ATCC, the company that sell Calu-1 told me that there is paper claiming about a 0.078 gene frequency for HLA A10 and a 15% phenotype in the Caucasian race back in 1976. In this case do you think that it would be feasible to screen non small cell lung cancer patients for HLA? From your experience from a statistical point of view, how many patients do you think that I will need to screen?
Thank you,
Alessia
You can find the updated allele frequencies for the various HLA Alleles here (from NMDP in the US):
https://bioinformatics.bethematchclinical.org/hla-resources/haplotype-frequencies/high-resolution-hla-alleles-and-haplotypes-in-the-us-population
From there you can use the binomial theorem to calculate the probability of zero "successes" (i.e. HLA matches in your sample) across a certain number of "trials" (i.e. samples to screen): http://stattrek.com/online-calculator/binomial.aspx
-Benjamin
Hi Benjamin,
These links are very useful! Is it HLA0101g a subtype of HLA10?
Thank you!
Alessia
I would guess that "A10" may be indicating a lower resolution typing than "A0101". In this case A0101g would not be a subtype of A10.
See this site for information re HLA nomenclature and typing resolution:
http://hla.alleles.org/nomenclature/naming.html
-Benjamin
Allesia- Did you ever resolve this issue- I am facing a similar challenge now, how did you end up proceeding?
Hi Rachel, sorry I missed your message. I didn't continue that project but you can try to contact MIDGAM (www.midgam.org.il)
Hi All, I am not an immunologist but I have a question with regards to the relevance of HLA class II in an experimental system when one tries to match a cancer cell line to alogeneic PBMCs. Why would MHC class II molecules matter if there are no professional antigen presenting cells among the cancer cells? Shouldn't you match only MHC class I for such ex vivo models? Thanks!
- Badges
- Science topic
- Similar topics
- Biological Science
- Histology