Monday, December 20, 2010

Snip, snip, the party's over? (Musings on the seductive rise of DNA typing of blood groups)

Updated: 28 Jan. 2017 (Fixed broken links)
December's blog's title comes from
  • A song by Willie Nelson: The party's over
  • Single-nucleotide polymorphism or SNP (pronounced snip), the basis of most blood group DNA typing. Simply put, a SNP is a DNA sequence variation that occurs when a single nucleotide in the genome differs between members of a species or paired chromosomes in an individual.
The use of seductive is tongue in cheek. Not sure if I mean 'harmful but enticing'. Perhaps insidious would be apter, i.e., 'developing so gradually as to be well established before becoming apparent'?

The blog discusses
  • Increasingly cozy relationship between TM professionals and the biotechnology industry;
  • Molecular typing of blood groups and its uses;
  • What DNA typing potentially means for the TM laboratory's work force;
  • A futuristic farce on where all this might lead.

The idea for this blog came from these sources:
  1. Announcement in the AABB Weekly Report (17 Dec. 2010) of a new section in Transfusion devoted to blood group genomics with Connie Westhoff, currently director of genomics and immunohematology at the New York Blood Center, as the section's associate editor.
  2. DNA typing resonated because a recent blog (Aug. 2010) discussed Immucor' s investment in DNA typing, specifically its purchase of Bioarray Solutions in Aug. 2008: Goldfinger's filings, a customer's toolkit (Musings on business intelligence) and Immucor's belief that "molecular immunohematology will revolutionize blood bank operations."
  3. By chance, in researching a case study, I had just read two articles on molecular typing, one co-authored by Westhoff:
Westhoff CM, Sloan SR. Molecular genotyping in transfusion medicine. Clin Chem 2008;54(12): 1948-50.
Reid ME. Transfusion in the age of molecular diagnostics. Hematology 2009.

FYI: Marion Reid, a legend at the New York Blood Center and recipient of the 2006 International Woman in Transfusion Award.
Interestingly, the Westhoff and Reid papers above list author involvement with Bioarray Solutions (and its earlier name, Bioarray), a company whose products DNA type red cell and HLA antigens.
Westhoff and Sloan paper's disclosure of potential conflicts of interests:Consultant or Advisory Role: C.M. Westhoff, BioArray, Ltd. and Immucor Inc. S.R. Sloan, Bioarray Solutions.

Reid paper's conflict-of-interest disclosure: The author receives research funding and royalties from patents from BioArray Solutions.
As evidenced by these examples, world class TM professionals often advise biotechnology companies and Bioarray Solutions has its share of respected advisors. Besides the three already mentioned, I came across others. It's accepted that industry needs input from professionals and the practice is common.

Of note, industry financing involving direct and indirect payment (consulting, food, travel, product samples, you name it) is estimated as high as $29 billion/year.

Questions to ponder:
  • Could consulting for a biotech firm, presumably for a fee, influence a person's perspective towards the technology in general?
  • How about earning royalties for a patent used by the company?
  • Could association with a company, current or past, influence a person in a role as editor of a journal section specifically dealing with research and developments in the company's field?
What happens when respected, leading experts, individuals who are the targetted clients of a particulate technology, also advise the companies as consultants and later serve as gate keepers of peer review for what is often company-sponsored research?

My gut tells me that something could happen, despite the obvious integrity of those involved. Arrangements, many of which are not transparent, seem too cosy to generate confidence in the system.

Can anything be done? Perhaps not. 

To quote from Immucor's 2010 SEC Form 10-K report in discussing Bioarray Solution's DNA technology:
"In many countries, blood pre-transfusion testing is limited to the prevention of transfusion reactions and not for the prevention of alloimmunization, which occurs when antigens foreign to the patient are inadvertently introduced into the patient’s blood system through transfusions. If alloimmunization occurs, the patient develops new antibodies in response to the foreign antigens, thereby complicating future transfusions.
By using multiplex, cost-effective molecular testing, our molecular technology allows testing to prevent alloimmunization for better patient care."
Not unsurprisingly Immucor focuses on preventing alloimmunization, as this use would considerably expand profits. As I noted in the earlier blog: "In a consumer society, if a real need does not exist, companies try to create one."

So what are the legitimate uses of molecular typing to determine blood groups?

Scenarios where DNA typing could prove useful include these scenarios (see Further Reading for more details of each scenario):

1. Recently transfused patients with unexpected antibodies who have two red cell populations.

2. Patients with a positive direct antiglobulin test where available typing antisera reacts only by the indirect antiglobulin test.

3. Fathers of infants carried by Rh negative woman with anti-D where the father is Rh positive.
DNA typing the father for the RhD gene can determine homo- or heterozygosity for D and indicate if the fetus is definitely D+ and should be monitored.
If being D-negative is possible, the fetus should be Rh typed using fetal DNA from cells obtained by amniocentesis or by testing cell-free, fetal-derived DNA in maternal plasma.

If the fetus is Rh negative, mother and fetus do not need to be monitored.
4. Pregnant women who appear to be weak D, in order to differentiate between weak D and partial D, since usually only the latter can make anti-D. Potentially, Rh immune globulin could be given only to females with a D variant who are partial D.

5. Extended phenotype matching of donor blood for patients with sickle cell disease.

The above scenarios are not exactly pressing. Many situations have work-arounds or present limited safety risks to patients. Granted they offer useful adjuncts to serologic testing.

As to the nuts and bolts of pretransfusion compatibility testing, routine ABO and Rh typing do not warrant molecular methods, nor does preventing alloimmunization in general as Immucor would have its investors believe.

Although molecular methods for typing red cell antigens are available in some locations, today the technology is mainly used by reference laboratories to supplement traditional serologic typing methods done 'in the trenches'.

Despite good progress, significant practical and scientific limitations remain and the issues are complex, as discussed in the papers in Further Reading. Seltsama and Doescher cover the scientific issues extensively. The 'high-throughput platforms' mentioned by adherents are not quite ready for prime time live in your neighbourhood blood centre and transfusion service.

Thus, work force effects will not be snip-snip, where SNIP is defined as
  • To cut, clip, or separate with short, quick strokes
Rather the SNP cut will be long and slow before staff hemorrhaging begins. But begin it will.

All TM laboratory professionals know that serology is in its last gasps. This is shown by the paucity of immunohematology papers in journals such as Transfusion. For example, December's issue has two and neither deals with red cells.

Rest assured that once 'high-throughput platforms' become widely available for molecular typing, regardless of whether sound scientific rationales exist, antigen phenotyping of all kinds will be taken over by molecular methods. It's simply too seductive, especially with so many TM leaders on side with the program.

As with automation, vendors will justify initial costs or high ongoing maintenance and supply charges (if a 'razor and blade' business model is chosen) with the promise of "reduced headcounts"  -  eliminating jobs for people in the community in favour of sending money to commercial enterprises elsewhere, where corporate profits are the main concern.

Seeing as it's the festive season, let's peek into a futuristic world populated by sugar plum fairies. I've mined some stereotypes (hope they bring a chuckle) and all is written with warm goodwill.

Here's the premise. What if companies pushing automation, other high throughput instruments, and information systems to support them have their way and almost every transfusion service laboratory job is eliminated?

In such a fairyland:

DNA typing platforms will match patients with donors for all major blood group antigens and pretransfusion serologic tests will become redundant.

Labs will be staffed by inexpensive 'multi-skilled workers' with no formal qualifications, all supervised by technical specialists (perhaps one per shift) now in a 'leadership role better aligned with their skills'.

Lab technologists / scientists will be free to take a buyout, a generous offer of $100 per year worked. They will, however, be forced to hawk their collection of syringes and tourniquets on street corners to finance a nursing education, which they can now consider since nurses no longer have much patient contact.

Since gowns became de rigeur for personal protection in the lab, they will sport white lab coats and happily be mistaken for physicians who have long claimed the lab coat as their uniform of record despite most never having set foot in a lab (too dark and dingy down in those basements). 

On the clinical side, other technology (bar code scanners, RFID, and the like) will ensure positive patient and donor unit identification. Smiling robots will administer blood transfusions, leaving nurses free to rant about newbie residents and those nitpicking idiots in the lab, have wild sex in the closets, and fantasize about Nurse Jackie one day killing Dr. House.

Yet more technology in the form of biosensors implanted in patients will decide when blood transfusion is warranted using programmed best practice guidelines, determine which components are needed, measure their effectiveness post-transfusion, and communicate all via wifi to a central information system nicknamed Big Bro (BB for short).

TS medical directors will sit in their offices playing on iPads with100s of fun apps all tailored to what Gregory House would be like as a hematologist versus a hematopathologist.
  • On breaks, the wunderkind millennials will tweet each other about how to pay off student loans before retiring.
  • About-to-retire boomers will obsess about stock portfolios.
  • Late Generation Xers will text about how, if they hear once more how great the Beatles and 1960s were, they will throttle their parents.
In the absence of sugar plum fairies, when reduced serologic antigen phenotyping eventually occurs, the serologic mainstays will remain as they are now, antibody detection and identification. Thank goodness for the immune response. The party's not quite over...."These are still the days, my friends".
Have a well deserved, fun-filled and relaxing holiday season with friends and family. And may the farce be with you....


Reid ME. The Rh antigen D: a review for clinicians. Blood Bulletin 2008 Apr; 10(1).

Reid ME. Transfusion in the age of molecular diagnostics. Hematology 2009.

Seltsama A, Doescherb A. Sequence-based typing of human blood groups. Transfus Med Hemother. 2009; 36(3): 204–212. Published online 2009 May 14.

Westhoff CM, Sloan SR. Molecular genotyping in transfusion medicine. Clin Chem 2008;54(12): 1948.

As always the views are mine alone. Comments are most welcome BUT, due to excessive spam, please e-mail me personally or use the address in the newsletter notice.