“Proof of Concept” Pilot Study: Bioprocess Chain of Custody and Bioresource Sample Management Temperature Observations. Sample Level Temperature Trends and Stability Data Obtained Via Utilization of Bluechiip ® Temperature Tracking Technology

Abstract

Background:

Preservation and optimization of biosample integrity to foster relevant research results and outcomes is a guiding principle of sample management. Tracking pre-analytical biospecimen lifecycle variables and bioprocessing chain of custody data enables documentation of adherence to best, regulatory and quality biobanking practices. Knowledge of individual sample and sample set temperature variability is believed to enhance delineation of artifacts during downstream analysis. Analysis of temperature responses may elucidate understanding of temperature trends which can aid downstream interpretation and provide an empirical foundation for “fit for purpose” sample management protocols and evidence-based biobanking practices. Bluechiip and the American Type Culture Collection (ATCC) conducted a pilot to test the bluechiip technology^® performance and validate key proofs of concept for tracking temperature of biological samples.

Methods:

One hundred six (106) Corning^® cryovials with bluechiip^® buttons and one hundred six (106) standard Corning^® cryovials labeled with 1-dimensional (1D) barcoded labels were evaluated. Identifiers were tracked and temperature data recorded in corresponding environments ranging from −192°C to +57°.

Results:

Nine of ten proof of concepts, defined in collaboration with ATCC successfully demonstrated functional capabilities of the bluechiip^® technology. Bar-code label read performance was compared, producing evidence demonstrating a high rate of failure on the bar-code arm.

Conclusion:

Temperature data collected heightened observations of sample temperature variability. Prevalence of bar-code label read failure and issues affecting reliability of barcode performance may be under-reported and unrecognized in sample management practice, particularly when the temperatures are lower than −60°C. It appears the bluechiip^® tracking technology may offer increased reliability over one-dimensional (1D) bar-coding technology; however while promising these findings require validation in future trials, including two-dimensional (2D) bar-coding technologies.

Introduction

Preservation and optimization of biosample integrity to foster relevant research results and outcomes is a guiding principle of sample management.¹ Tracking pre-analytical biospecimen lifecycle variables and bioprocessing chain of custody data enables documentation of adherence to best, regulatory and quality biobanking practices. Knowledge of individual sample and sample set temperature variability is believed to enhance delineation of artifacts during downstream analysis.¹ Comprehension of sample temperature response 1) bestows capability for real time acquisition of empirical data to facilitate understanding of temperature trends to aide downstream interpretation and 2) permits introduction of “fit for purpose” sample management practice. Aggregated annotation of sample temperature may afford insight into the effect of environment and bioprocessing on sample level temperature. Tracking temperature observations over time may also allow one to predict sample temperature responses, actualizing real time implementation of evidence based biobanking practice (EBBP).²

Until recent technological developments, specifically the bluechiip^® technology (a passive wireless technology based on MEMS (micro electromechanical systems)) it was not possible to obtain sample level temperature or assess the degree of individual sample and sample set temperature variability. Piloting of emerging technologies is considered a best practice.³ Furthermore, one can intimate that products with user centered design are likely to have higher rates of utility and reliability.⁴ Therefore, Bluechiip and the American Tissue Culture Collection (ATCC) conducted a pilot to test and validate the bluechiip^® technology's performance. Ten proofs of concept (Table 1) were defined. Pilot objectives are noted in Table 2.

Table 1.

Pilot Proof of Concept Requirements

1	Compatibility with bioresource workflow and bioprocesses
2	Integration into consumables
3	Real time identification of cryovials
4	Real time identification of unique identifier
5	Support chain of custody
6	Capture temperature over critical lifecycle time points
7	Function in a cryogenic environment
8	Withstand, operate and perform in a frost environment
9	Function in a wide range of temperatures
10	Survive and operate post gamma radiation

Table 2.

Pilot Primary Objectives

1	Demonstrate that an individual sample or set of samples can be accurately interrogated in real-time using the bluechiip^® Matchbox™ reader to support cell culture and sample management (QMS) practices.
2	Demonstrate that the tag measures the range of temperatures specified.
3	Specimen biosketch can be formed from time and temperature history measurements.
4	Achieve reliable operation of the bluechiip^® tags at −80°C to −196 °C and temperatures relevant to cell culture based biopreservation, specimen processing, transport & cryogenic storage
5	Correlate environmental temperature measurements using the bluechiip^® technology
6	Acquire data for accurate temperature graphs of the samples
7	Ability to track through relevant portions of the process chain without missing a location
8	Completeness and accuracy of data obtained/entered into the database

Materials and Methods

Background

Proof of concept pilot tests were conducted over four days including length of storage readings taken at 16, 104 (∼3 months) and 284 days (∼9.5 months). Products tested were the bluechiip^® button ID device embedded into a Corning^® 1.2 ml cryovial (Fig. 1) and the bluechiip^® Matchbox™ reader (see Fig. 3). The pilot evaluated the bluechiip^® technology's performance vs. bar-code scanning across thirteen trials (Table 3) relevant to repository and laboratory sample management processes. Sample sizes of each trial are listed in Table 3. ID and temperature sensing capability was evaluated for samples exposed to temperatures of +37°C to −196°C (Table 4). Structural integrity of the vials was evaluated after exposure to temperatures between +121°C to −196°C. Cryovial identifiers (ID) and temperature were ascertained by removing the sample briefly from its corresponding environment and placing the cryovial directly on the Matchbox™ reader. Chain of custody and biospecimen lifecycle interval temperatures (Table 5) were measured for the bluechiip^® arm. ID was tracked for the barcode arm. Sample temperature history was acquired by the Matchbox™ reader when the cryovial was read by the reader. Corresponding date and time were recorded in the STREAM™ database. All tests were conducted on-site at the ATCC biological resource facility in Manassas, Virginia.

FIG. 1.

Bluechiip tagged Corning cryovial.

Table 3.

Sample Size per Trial

Trial	Bluechiip^®	Barcode
Cell Culture Cryo-preservation	40 (+probe)	40
Post Gamma Radiation	106	0
Freeze/Thaw	44	46
Centrifuge	5, 15, 30, 50	5, 15, 30, 50
Snap-freezing	62	43
Frost	52	46
Autoclaving	6	6
Microwave	6	6
Drop test	6	6
Day 2 Retrieval	53	44
LOS Retrieval (16 days)	54	74
LOS Retrieval (≈3 months)	53	74
LOS Retrieval (≈9.5 months)	53	72

Table 4.

Range of Temperatures Assessed per Trial

Trial	Range of temperatures assessed (°C)
Cell Culture Cryo-preservation	−178 to +15
Post Gamma Radiation	+15 to +23 (cell culture)
	+17 to +22 (Field trial)
Freeze/thaw	−192 to +11
Centrifuge	−4 to +35
Snap-freezing	−10 to −125
Frost	−47 to −123
Autoclaving	+23 to +57
Microwave	+10 to +34
Drop test	−3 to +15
Day 2 Retrieval	−52 to −178
Length of Storage Retrieval (Day 16)	−90 to −173
Length of Storage Retrieval (Day 104)	−81 to −169
Length of Storage Retrieval (Day 284)	−80 to −175

Table 5.

Chain of Custody Lifecycle and Temperature Time Points Measured per Trial

Pilot Trials (10 types)	Temperature Measurement Time points
Post gamma radiation scan	Post gamma irradiation at Time 0
Cell culture preservation workflow exercise	Baseline
	Time of filling/pre-preservation
	Post-preservation/time of transport
	Time of storage
	Day 2 Retrieval from storage
	Day 16 Retrieval from storage
	≈3 months (Day 104) retrieval from storage
	≈9.5 months (Day 284) retrieval from storage
Freeze-thaw field trial	Baseline/Time of Retrieval
	Time of transfer to water bath
	Post-thaw/removal from water bath
Centrifugation field trial	Pre and Post centrifugation runs
Snap-freeze field trial	Post snap-freeze
Drop test	Pre and post drop runs
Microwave test	Pre and post microwaving
Frost field trial	Post frost
Autoclaving field trial	Pre and post autoclaving
Length of Storage field trials	Time of Storage (Day 0)
	LOS Retrieval (Day 2)
	LOS Retrieval (Day 16)
	LOS Retrieval (Day 104)
	LOS Retrieval (Day 284)

Trial set-up

On-site at Bluechiip Headquarters in Melbourne, Australia, bluechiip^® tags were inserted into the base of cryovials using an overmoulded plastic insert or “button”. “Buttons” were constructed so that cryovial functionality (de-capping lock and height) was not affected. (The tag was not in contact with the sample, but was in contact with the inside surface of the cryovial.) Immediately following, bluechiip^® embedded Corning^® 1.2 ml cryovials were irradiated with 25kGy. Pre-trial, the temperature behavior of a number of tags was measured against a platinum thermocouple. From this data an algorithm was determined to calculate the temperature from a measure change in received signal. The tags were then calibrated on-site at ATCC.

Unique bluechiip^® tag identifiers (ID) and/or 9 character 1-dimensional (1D) barcode identifiers were assigned to each cryovial prior to the trial, enabling comparison between the two technologies. Corning^® 1.2 ml cryovials were used for the barcode arm printed on standard thermal labels. Total sample size on each arm was 106. The number of cryovials used for the primary trials was 96. During the trials, sub-sets from the 96 cryovials were split for workflow and field trials. Ten additional cryovials were sequestered for microwave and drop testing trials.

Prior to commencement of testing, Bluechiip^® cryovials were read to confirm baseline ID and temperature. Prospectively for all trials, Bluechiip^® cryovials were scanned to confirm ID, track chain-of-custody and ascertain sample level temperature; barcode cryovials were scanned to confirm ID (Table 5).

Conditions for “failure” were established and tracked. A “failed” read on the bluechiip arm was defined as a cryovial that could not be read when placed in the reader. A “failed” read on the barcode arm was “the need to wipe surface area” to induce a successful read. If a Bluechiip^® tag completely failed, as in the drop test it could not be read again. (The temperature sensing and identification function of the tags are integrated and not separate, hence ID and temperature could be not be measured and vice and versa.) However, if a bar-code cryovial “failed” e.g. due to frosting it could be read again. Failed cryovials were not removed from the sample set, to maintain sample size. Partially failed tags were carried through and as a result additional instances of failure were recorded.

Data, logs and monitoring

Trial read duration was obtained for both arms throughout all processes. To support validation of temperature readings, environmental temperature measurements were obtained and recorded at key intervals, including pre-trial and post-trial. Instrumentation thermostat and probe vial readings were recorded when available.

Gamma radiation trial

bluechiip^® cryovials were tested for ability to survive and function following gamma irradiation. This was determined by successful read upon removal from the study kit.

Cell culture cryopreservation trial

Labeled cryovials were filled with 1.0 ml of ATCC standard media at room temperature. Cryovials were scanned, attached to individually identified canes and placed in a Thermo^® controlled rate freezer (CRF) for preservation time of ninety minutes. CRF temperature was tracked and correlated to the internal monitor and probe cryovial. Upon completion, the canes were transferred to a cryo-bin containing liquid nitrogen (LN2). Environmental temperatures were recorded and the bluechiip^® cryopreserved samples were scanned. Bar-code cryovials were not read at this time. Canes were distributed evenly in two tiers of a MVE^® 1800 Series cryo-tank in ATCC's biorepository.

Freeze/thaw trial

Cryovials were removed from cryo-tank, placed into a cryo-bin, scanned and transferred to the laboratory. Cane read time was documented and the samples were thawed in a +37°C water bath for seven minutes before scanning.

Centrifugation trials

Thawed samples were spun at 280 g for eleven minutes, at +4°C, repeating with four runs of progressively increasing samples sizes. Cryovials were scanned pre and post centrifugation.

Autoclave trial

Ten samples (five per arm), were placed in the autoclave for fifty-nine minutes. The log, documenting gravity, PSI and sterilization temperature, was obtained on completion. Samples were scanned pre and post autoclaving.

Snap freeze trial

Cryovials were scanned, placed on canes and submersed in LN2 in a covered cryo-bin. Samples were snap-frozen for forty-two minutes, pulled from submersion and read immediately. The majority of the cryovials remained on canes during the submersion; a minority were displaced and remained immersed throughout scanning.

Drop test

Twelve emptied cryovials (six per arm) were scanned and dropped from a height of two meters onto a concrete floor. Two runs were performed. Cryovials were scanned after each drop.

Microwave test

Twelve emptied cryovials (six per arm) were scanned and placed in a domestic microwave for five minutes. Cryovials were removed, inspected for structural damage and scanned.

Frost trial

Frozen, filled cryovials were scanned and placed in the vapor phase in a cryo-bin until an adequate level of frost was created. Cryovials remained in the cryo-bin (Fig. 2) while post-frost scans were performed.

FIG. 2.

Cryovials in cryo-bin during frost trial.

Length of storage (LOS) studies

Cryovial ID, temperature measurements and bluechiip^® technology cryogenic survival were assessed at Days two, sixteen and after approximately three and nine and a half months (104 and 284 days). Cryovials were removed from the two tiers of the LN2 tank and scanned to evaluate survival. After reading, all samples were returned to previous storage locations.

Trial analysis

Post pilot database logs underwent quality review of data. Individual sample temperatures and identifiers were confirmed for the bluechiip^® arm for each trial. Sample and sample set temperature variability were examined across trials for the bluechiip^® arm using reference tables and histograms to observe temperature trends over time and distribution of temperature across sample sets. Post pilot engineering analyzed the temperatures obtained against the algorithm defined. It was not possible to confirm accuracy of the bar-code ID measurements. Utility factors i.e. read rate duration and prevalence of failure were calculated for both arms.

Results

Proofs of concept

Nine of ten proofs of concept (Table I) across 13 tests (Table III) were successful. Confirmation of unique identifiers was not demonstrated in real time.

• Compatibility with bioresource workflow and bioprocesses: The technology demonstrated compatibility for support of sample level temperature and ID lifecycle tracking.

• Integration into consumables: The tag, button and cryovial integration remained intact through all processes, even with extreme processing. There was a minor expected cosmetic alteration in vial plasticity during autoclaving but the technology and cryovial retained functionality.

• Real time identification of cryovials: There were no observed errors in identification reads in the bluechiip^® arm throughout the duration of the trial. One error in identification occurred on the bar-code arm during the 16 day LOS trial.

• Real time identification of unique identifier: It was not possible to verify unique identifiers during the execution of the trial. Post-trial, a quality control review of the data was performed to check for any unique identification errors. There were no inaccuracies in the database.

• Support chain of custody: The bluechiip^® system demonstrated capability to support temperature and chain-of-custody tracking from acquisition to time of distribution.

• Capture temperature at critical lifecycle time points: The bluechiip^® technology demonstrated ability to capture temperature readings over relevant time points throughout the specimen lifecycle and defined relevant temperature intervals (+37°C, +4°C, −20°C −80°C, −148°C, and −196°C).

• Function in cryogenic environment: The bluechiip^® technology performed consistently while immersed in liquid and vapor phase LN2. Barcode arm failure rates were more prevalent than bluechiip^® tag reads. Barcode failure rates were 48.8% following snap-freezing and 31.8% at Day 2 Retrieval. More noticeably, failure rates progressively increased over time with long term storage, 16.2% after 16 days, 41.9% after 3 months and 73.6% after nine months.

• Withstand, operate and perform in a frost environment: The bluechiip^® cryovials and Matchbox™ reader consistently performed with significant frost. Cryovials did not have to be wiped to be scanned on the bluechiip^® arm (Fig. 3). There was an 87% failure rate on the bar-code arm.

• Function in a wide range of temperatures: The trial demonstrated the bluechiip^® technology can measure temperatures up to +57°C and down to −192°C. The trial also demonstrated the technology's ability to withstand and function after exposure to +121°C. While the barcode scanner did operate across temperatures, data demonstrated the functionality of barcodes declined around −60°C and more significantly below −80°C. The bar-code arm demonstrated prevalent failure (45.7%) in reads taken post thaw.

• High temperature point (+): The two highest temperature measurements recorded were +57°C (post autoclaving) and +34°C (post microwaving).

• Low temperature points (−): The two lowest temperature measurements recorded were −192°C (pre-thaw time of removal from the cryo-bin for transfer to the water bath) and −178°C (during Day 2 LOS).

• Survive and operate post gamma radiation: 5/106 cryovials failed post gamma radiation. The 4.7% failure rate (Table 6) prior to start of trial was determined to be a mechanical issue with the pre-production chips used for testing and was not related to the radiation.

FIG. 3.

Frosted tagged cryovial on Matchbox^™ reader.

Table 6.

Incidence and Prevalence of Failures per Pilot Field Trial

Trial	Bluechiip^® Arm Failures		Barcode Arm Failures
Post Gamma Radiation	5/106	4.7%	N/A	-
Cell CultureCryopreservation	1/40	2.5%	N/A	-
Freeze/Thaw	6/44	13.6%	21/46	45.7%
Centrifuge	1/50	2.0%	0/50	0%
Snap-freezing	3/62	4.8%	21/43	48.8%
Frost	4/52	7.6%	40/46	87.0%
Autoclaving	0/6	0%	0/6	0%
Microwave	0/6	0%	0/6	0%
Drop test	5/6	83.3%	0/6	0%
Day 2 LOS retrieval	5/53	9.4%	14/44	31.8%
Day 16 LOS retrieval	1/54	1.8%	12/74	16.2%
Day 104 LOS retrieval	0/53	0%	19/62¹	30.6%
Day 284 LOS retrieval	0/53	0%	22/41	53.7%

11/106 vials failed in total throughout the trial. Failed vials were not removed from sample set.

Temperature observations

Pre-analytical temperature variation at the individual sample level was positively observed across all sample sets during bioprocessing, sample handling and exposure to differing conditions. Throughout the pilot, differences in stability of sample temperature in lower and higher cryogenic environments were observed. Large ranges in temperature variability were observed during cryogenic processing and when samples were immersed into or removed from cryogenic environments.

Duration of reads

The average successive read rate was five seconds. Bluechiip^® arm read-rate durations were longer than the barcode arm. However, in instances when barcode read function was failing, barcode read-rate durations did increase. Bluechiip^® arm read functionality was consistent across all time points and temperatures.

Reliability and incidence of failures

Operative reliability was not 100% for either arm of the trial (Table 6). Eleven total bluechiip^® cryovials failed throughout the trial with a total failure rate ranging from 4.7% (5/106) (excluding drop test) to 10.4% (11/106) (including drop-test). It was not possible to track the total number of bar-code cyrovials that failed. It is possible, that Bluechiip^® tag failure prevalence was due to the decision to include pre-production tags out of specification due to manufacturing related time constraints, however, the authors cannot speculate further at this time. The bluechiip^® button cryovial concept has been re-designed to withstand drops and shocks. Per trial, reliability appeared to be more of an issue on the barcode arm, where barcode failures were significantly prevalent. Prevalence of failures was significantly greater in frequency across the barcode arm for four of the eight tests conducted. While prevalence of barcode failure was 0% for three of the tests (centrifuge, autoclaving and microwaving), reliability ranged between 16.2–87% for the other tests (Day 2 LOS (31.8%), Day 16 LOS (16.2%), Day 104 LOS (41.9%), Day 284 LOS (73.6%), post-thaw (45.7%), snap-freezing (48.8%) and frost (87%)). Greater incidence of failure was observed on the barcode arm than on the bluechiip^® arm in every trial except for the centrifuge and drop test.

Discussion

The bluechiip^® technology performed reasonably well across the process chain for extended periods of time in cryogenic and non-cryogenic environments as demonstrated by cell culture workflow and field trials (+56°C to −173°C). Pilot testing enabled comparison to bar-code read performance, which demonstrated high rate of failure. The high incidence of barcode arm failure (up to 87%) compared to the bluechiip^® arm up to 13.6% (excluding the drop test), while generally expected was surprising to observe specific to distinct temperatures and lifecycle time points. It can be hypothesized that bar-code failure prevalence rate and issues affecting reliability of barcode performance may be under-reported and unrecognized in sample management practice. Barcode technology was observed failing when exposed to temperatures outside the cryogenic space, as low as −60°C. Incidence of barcode read failure increased significantly when temperatures exceeded −80°C. Failure rates varied amongst sample processes, with increased prevalence at: snap-freeze, frost and when transferred in and out of cryogenic temperatures.

While post pilot analysis confirmed that the variation in responses from different tags appears to support the accuracy of the temperature measurements obtained, it is premature to comment further. Regardless, the ability to collect individual temperature data heightened observations of sample temperature variability. Temperature tracking data may enhance understanding of sample temperature response behavior, temperature trends over time and factors that affect sample temperature variability.

Empirical data collected during the trial appeared to support the following conclusions:

1. Distinct fluctuations in sample temperature occur across the specimen lifecycle.

2. Prevalent and consistent temperature variation appears to exist in individual samples and sample sets during sample management, despite standardized processing and environmental exposure.

3. Remarkable variation exists with samples in close proximity to one another (e.g. different temperatures were recorded for neighboring cryovials stored on canes).

4. Variability of sample temperature may be affected and impacted differently based on type of bioprocessing, environmental conditions, and degree of sample handling, mode, format and duration of biostorage.

5. Samples stored at the same temperature do not appear to maintain the same temperature as their corresponding sample set in during transport, short or long term biostorage.

6. Stability of temperature may be directly linked to length of storage and range of cryogenic temperature. Samples stored for durations longer than 2 weeks in temperatures colder than −150°C appeared to be more stable when exposed to room air and lower cryogenic conditions than samples stored for 2 days in the same conditions.

One could intimate that 1-D barcode labeling may be a less reliable method of tracking ID for samples managed in cold chain (−60 to −80°C) and cryogenic temperatures (below −80°C) than bluechiip^® technology due to poor performance observed during the trial. However while the results obtained are promising, this conclusion may be premature based on the fact that these findings require validation in future trials and acceptable parameters for reliability need to be confirmed with end-users. Demonstration of real time confirmation of unique identifiers is also required to support best practice. Furthermore, additional studies testing expanded sample sizes, and comparing performance against advanced technologies (i.e. 2-dimensional barcodes) in concurrence with operational trials are recommended. Such studies would serve to confirm the aforementioned hypothesis regarding reliability of performance and increase statistical significance of results obtained. Ongoing testing should aim to demonstrate product functionality and length of storage survival beyond 9.5 months.

Footnotes

Acknowledgments

The authors wish to thank Corning Life Sciences (Dawn Jackson and Mark Rothenberg) for their collaboration, contributions to sponsorship of the pilot and time; ATCC Management and Staff (Stewart Davis, Ken Jones, Susan Carlson, Doug Daley, Sherill Smallwood and Jody Terry)for their collaboration and assistance in conducting the pilot; and Bluechiip Engineering and Informatics team members (David Beard, Rick Zheng, and Shane Harper) for their product development contributions and time preparing the technology for the pilot.

Author Disclosure

Lisa B Miranda/Biobusiness Consulting, Incorporated discloses that the majority of work on this project was performed as a compensated consultant/contractor of Bluechiip Limited. Any remaining work not funded was performed in collaboration with Bluechiip and ATCC.

References

National Institutes of Health. National Cancer Institute's Biorepositories and Biospecimens Research Branch, (BBRB) Best Practices. 2011. http://biospecimens.cancer.gov/practices/2011bp.asp.

Miranda

“Obtaining and Maintaining High-Quality Tissue Samples: Scientific and Technical Considerations to Promote Evidence-Based Biobanking Practice (EBBP).”

Wigglesworth

et al. Management of Chemical and Biological Samples for Screening Applications: Wiley, 2012; 59–79.

International Society for Biological and Environmental Biorepositories (ISBER) 2012 Best Practices for Biorepositories: Collection, Storage Retrieval and Distribution of Biological Materials for Research. Biopreservation and Biobanking, 2012; 10,2:79–161.

Dillon

et al. User Acceptance of New Information Technology: Theories and Models. Annual Review of Information Science and Technology, 1996; 31:3–32.