INFORME SOBRE LOS EFECTOS DE LAS ONDAS DE RADIOFRECUENCIA

QUE EMITEN LOS SISTEMAS RFID EN PRODUCTOS FARMACÉUTICOS,

SANITARIOS Y BIOLÓGICOS.

Yuresky Rojas Rincón1 Javier Muñoz Giner2, 1Máster (Universidad Politécnica de Valencia - España). Ingeniera de Telecomunicaciones

(Universidad Santo Tomás – Colombia). yuresky@tagingenieros.com 2Doctorado (Universidad Politécnica de Valencia - España). Ingeniero Telecomunicaciones

(Universidad Politécnica de Cataluña - España). Director General TAG Ingenieros. javier@tagingenieros.com

Abstract - This documents presents a report based on the variety of studies that has been made regarding to the possible effects of the Radio Frequency Radiation on pharmaceuticals, medical and biological products. In this regard, the report concludes that further than cause any harm, RFID the adoption of radiofrequency identification track and trace system may contribute significantly to public health. Diferentes estudios se han hecho acerca del impacto potencial en la calidad de productos farmacéuticos y/o biológicos debido a la exposición prolongada a ondas RF (Radiofrecuencia), como ocurre en los casos en los que se adoptan los sistemas de Identificación por Radiofrecuencia -RFID, para realizar seguimiento y localización de este tipo de productos.

En este sentido, ha sido necesario realizar estudios que avalen la seguridad en el uso de este tipo de sistemas RF sobre los elementos más sensibles del sistema sanitario. Estos estudios tienen en cuenta el tiempo de exposición, potencia y diferentes frecuencias a las que trabaja.

Existe evidencia experimental de que los efectos de la exposición a radiaciones de RF de este tipo de productos son mínimos y despreciables en cuanto a que los cambios de temperatura que se producen no alteran de ninguna manera la calidad de las muestras; tanto así que la FDA (Food and Drugs Administration) promueve el uso de los sistemas RFID como un componente estratégico para darle mayor seguridad e integridad a los productos y así mismo a la seguridad de los pacientes [1].

Este informe se basa en las conclusiones de varios estudios realizados por diferentes instituciones de gran prestigio internacional que avalan la veracidad de los resultados obtenidos y demuestran la inocuidad de los sistemas RFID a cualquier tipo de frecuencia.

En un estudio reciente, el centro de sangre de Wisconsin y el laboratorio de RFID de la Universidad de Wisconsin han determinado que la aplicación de la Identificación por Radiofrecuencia tiene importantes implicaciones tecnológicas y orientadas a procesos, que hacen técnicamente viable y económicamente justificable incorporar el RFID en los procesos inmersos en los centros de sangre [3].

La evaluación de la tecnología RFID en los procesos de los centros de sangre implican la legibilidad de la etiqueta, la arquitectura y diseño de sistemas RFID, estándares, el desempeño, la temperatura y los efectos biológicos que las ondas de RF pueden tener sobre los productos sanguíneos.

La Identificación por Radiofrecuencia puede ayudar a superar una serie de retos comunes y las ineficiencias asociadas con el proceso de identificación y seguimiento de los productos sanguíneos. La tecnología RFID trabajando a alta frecuencia (HF - 13,56MHz) se desempeña de manera adecuada y segura para los glóbulos rojos y las plaquetas sin alterar de manera significativa sus características. Según el informe de la Universidad de Wisconsin, las mejoras en la calidad y productividad dadas por la implementación de la tecnología RFID en los diferentes procesos permiten que los costos de inversión puedan recuperarse en aproximadamente 4 años [3].

Otros estudios se han realizado para determinar la viabilidad de los sistemas RFID en frecuencias más altas como la UHF, para ello se han sometido los productos a escenarios extremos; estudios con resonancia magnética NMR (Nuclear Magnetic Resonance) en frecuencias de 800 -900 MHZ son válidos para RFID teniendo en cuenta que muchos sistemas de identificación por RF usan este rango de frecuencias. La exposición de RF en este rango no causa daños no térmicos a las bio-moléculas. Y eso, teniendo en cuenta que los productos farmacéuticos y sanitarios fueron expuestos a la Radiofrecuencia de sistemas NMR que es más de un millón de veces la radiación que puede ocurrir con la tecnología RFID.[1]

Para frecuencias de 915MHz y 2,45GHz se ha demostrado que la radiación RF (RFR) incrementa la temperatura pero de manera casi insignificante en muestras acuosas. [2]

Algunas conclusiones sobre el impacto potencial en la calidad de los productos cuando han sido expuestos hasta por 16 horas son:

o Incremento en la temperatura de 0,5 a 1 grado Celsius.

o No hay cambios en magnitud, pureza ni potencia.

o Según los estudios de estabilidad del ICH (International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use), la variación de la temperatura puede ser de alrededor de 2oC en cámara de temperatura, con lo cual la variación dada por la radiación RF de casi 1o se considera despreciable [4].

A continuación se muestran algunos casos de éxito de aplicación de la tecnología RFID a la industria en el sector socio-sanitario y algunos ejemplos de aplicación en empresas que suministran activos médicos a hospitales o centros médicos (bolsas de sangre, implantes ortopédicos, etc.).[6]

- Mallinckrodt Pharmaceuticals, fabricante de medicamentos genéricos, implementó en 2005 un sistema basado en RFID con el fin de mantener una vigilancia sobre los medicamentos a lo largo de la cadena de suministro. - Intelligentz Corporation ha desarrollado un sistema para eliminar las falsificaciones de medicamentos utilizando tecnología RFID. Una base de datos genera un código unívoco para cada pastilla, código que se envía a los fabricantes a través de Internet. - Zimmer, multinacional distribuidora de productos de cirugía ortopédica, ha instalado lectores RFID a través de sus centros operativos en Nueva Zelanda, Australia, Japón y Tailandia. Los implantes ortopédicos individuales forman parte de un kit más amplio que se suministra a hospitales y en escenarios de operaciones. La creación de los kits se lleva a cabo con un 100% de precisión, asegurando que todos los ítems son fácilmente visibles para facilitar la calidad del envío previo y su posterior recepción. Cuando esos kits son devueltos para su actualización, limpieza y acondicionamiento anual, es necesario revisarlos uno a uno, cuando únicamente el 3% de su contenido suele estar utilizado. Las pérdidas de tiempo y costes eran significativas hasta que el etiquetado de ítems mediante RFID ha acelerado dicha comprobación además de que la precisión en el registro de inventarios ha mejorado drásticamente. - Pfizer posee un piloto en este ámbito, con la utilización de RFID para evitar falsificaciones de la píldora Viagra. - Purdue Pharma integra etiquetas RFID en sus líneas de embalaje farmacéutico de alta velocidad con el objetivo de mejorar la eficiencia y la seguridad de la cadena de suministro farmacéutica. - Farmacéuticos MAYPO (distribuidor de medicamentos en México), ha inaugurado la primera línea de etiquetado de medicamentos con tecnología de identificación por radiofrecuencia (RFID) del país, que permite su autentificación e identificación única proporcionando gran trazabilidad a lo largo de los procesos logísticos. MAYPO apoya así el modelo de provisión de medicamentos presentado por la Comisión Nacional de Protección Social en Salud conocida como Seguro Popular, que integra la tecnología RFID para proveer las recetas electrónicas programadas en las tarjetas inteligentes de sus afiliados. - En EE.UU. la National Patient Safety Agency (NPSA) ha publicado sus recomendaciones relacionadas con los sistemas de trazabilidad electrónica para pacientes y bolsas de sangre, mediante una especificación llamada Electronic Clinical Transfusion Management System (ECTMS), que pretende gestionar la historia clínica de las transfusiones electrónicamente. El objetivo de estas especificaciones es garantizar que el paciente reciba la transfusión correcta. El sistema ECTMS necesitará integrar servidores centrales, ubicaciones clientes y terminales remotos de acuerdo a las especificaciones del organismo NPSA. La especificación cubre la trazabilidad automática de productos como las bolsas de sangre desde el punto de extracción a la transfusión final, además permite gestionar la administración de los activos de sangre.

Para finalizar, se puede inferir que los diferentes estudios demuestran que la exposición a energía por RF debido a la implementación de un sistema RFID de seguimiento y trazabilidad representa un riesgo que no es significativo en cuanto a los efectos que pueda tener al alterar la calidad de los productos farmacéuticos y sanitarios, ya sea de tipo molecular, biológico, físicos, medicamentos basados en proteínas, etc.

ANEXOS

ANEXO 1: A Method to Investigate Non-Thermal Effects of Radio Frequency Radiation on Pharmaceuticals with Relevance to RFID Technology

ANEXO 2: Report to FDA of PQRI RFID Working Group by Robert H. Seevers, Ph.D.

REFERENCIAS [1] Report to FDA of PQRI RFID Working Group.Robert H. Seevers, Ph.D. 2005 [2] A Method to Investigate Non-Thermal Effects of Radio Frequency Radiation on Pharmaceuticals with Relevance to RFID Technology. Felicia C.A.I. Cox, Vikas K. Sharma, Alexander M. Klibanov, Bae-Ian Wu Member, IEEE, Jin A. Kong Fellow, IEEE, and Daniel W. Engels Member, IEEE 2006 [3] Tracking blood products in blood centres using radio frequency identification: a comprehensive assessment. Rodeina Davis, Bradley Geiger, Alfonso Gutierrez, Julie Heaser, Dharmaraj Veeramani Vox Sanguinis Volume 97 , Issue 1 , Pages50 - 60 © 2009 International Society of Blood Transfusion

[4] Letter from Jamie T. Hintlian, Partner, Accenture LLP, to Dr. Paul Rudolph of FDA, 2004. [5] Liquid Pharmaceuticals and 915 MHz Radiofrequency Identification Systems, Worst-Case Heating and Induced Electric Fields, H Bassen, RFID Journal, 2005. [6] Informe de Vigilancia Tecnológica. Tecnología de Identificación por radiofrecuencia (RFID): aplicaciones en el ámbito de la salud. Javier I. Portillo, Ana Belén Bermejo, Ana M. Bernardos. 2008

ANEXO 1

Abstract—A method is reported to accurately and

precisely control temperature of a solution sample to

investigate non-thermal effects of radio frequency radiation

(RFR) on pharmaceuticals. This method utilizes a transverse

electromagnetic (TEM) cell connected in series with a

radiation source. The temperature of a sample under study,

within the TEM cell, is regulated using a combination of a

fiber-optic thermometer and thermo-electric cooler. It is

shown that the sample temperature can be accurately

controlled and maintained even under conditions where the

RFR can increase the sample temperature via thermal mode.

This methodology provides a well-controlled approach to

investigate the non-thermal effects of RFR for a range of

incident power intensities and frequencies and initial sample

temperatures.

I. INTRODUCTION

N its February 2004 report, the U.S. Food and Drug

Administration (FDA) listed Radio Frequency

Identification (RFID) technology as an important tool to

combat counterfeiting of pharmaceutical products [1].

This technology utilizes labeling of pharmaceutical

products with tags, which can be tracked using

electromagnetic radiation in the radio frequency range

(electromagnetic radiation in the frequency range of 3

KHz – 300 GHz) similar to those used in radio

communications, wireless data networks and mobile

phones. One of the concerns regarding successful

implementation of this technology, however, is whether

the RFR is capable of causing deleterious reactions in the

drug product leading to reduced potency and/or other

undesirable effects.

The effects of RFR on materials have been broadly

classified into two categories: thermal effects and non-

thermal effects [2]-[4]. Thermal effects are defined as

Manuscript received April 3, 2006. This work was supported in part

by the HRI/Pharmaceutical Industry Consortium at Auto-ID Labs,

Massachusetts Institute of Technology (MIT), Cambridge, MA 02139.

B.-I. Wu is with the Center for Electromagnetic Theory and

Applications (CETA) at MIT, Cambridge, MA 02139 USA (phone: 617-

253-4028; fax: 617-253-0987; e-mail: biwu@mit.edu).

J. A. Kong is head of CETA at MIT, Cambridge, MA 02139 (e-mail:

kong@emwave.org).

F. C. A. I. Cox is with Department of Electrical Engineering at MIT,

Cambridge, MA 02139 (e-mail: fcaic@mit.edu).

V. K. Sharma is with Department of Chemistry at MIT, Cambridge,

MA 02139 (e-mail: v_sharma@mit.edu).

A. M. Klibanov is with Department of Chemistry and Bioengineering

at MIT, Cambridge, MA 02139 (e-mail: Klibanov@mit.edu)

D. W. Engels is with Auto-ID labs at MIT, Cambridge, MA 02139 (e-

mail: dragon@csail.mit.edu).

those that stem from an appreciable increase in the

temperature of a substance on exposure to RFR, i.e., it is

this heating (measurable using an appropriate temperature

probe) that is responsible for the observed changes. From

the pharmaceutical point of view, it is reasonable to

assume that RFR-induced thermal effects would be no

different than those induced by conventional bulk heating,

for example, as produced by exposing a product to heat in

a controlled temperature oven or on the shelf in an

uncontrolled room environment. Since the effects of

temperature on drug stability are generally well

understood, the effects of RFR-induced thermal effects

can be easily predicted.

In contrast, non-thermal effects, those arising even in

the absence of any appreciable increase in the material

temperature upon exposure to RFR, are obscure and

controversial. Although, there have been a few studies of

non-thermal effects of RF radiation on biochemical

samples, such as cells and tissues, the conclusions from

these studies are inconsistent and provide no mechanistic

understanding of the non-thermal effects.

To properly assess the non-thermal effects of RFR, it

is important that the temperature of the sample under

study is precisely and accurately controlled and

maintained. This should be true even under conditions

where the RFR used would tend to elevate sample

temperature. In the few studies published thus so far on

evaluating the non-thermal effects of RFR, temperature

stability is usually accomplished by using a low frequency

RFR that is not absorbed by the sample or a low power

intensity of a high frequency RFR such that the

temperature increase is not appreciable [5],[6]. The overall

temperature control is obtained by keeping the test-

assembly in controlled conditions. However, no direct

means of temperature control is used to maintain precise

sample temperature in these prior studies. This approach

limits the use of a high power intensity of a high

frequency RFR (e.g., those in the microwave region) since

the temperature increase due to absorbed power would be

much greater and thermal effects would become

important. The reason for using above-mentioned

approaches to attain temperature control is presumably

due to the fact that materials commonly used for

temperature control such as metallic surfaces and liquids,

interfere with the absorption of RFR by the sample under

consideration.

AMethod to Investigate Non-Thermal Effects of Radio Frequency

Radiation on Pharmaceuticals with Relevance to RFID Technology

Felicia C.A.I. Cox, Vikas K. Sharma, Alexander M. Klibanov, Bae-Ian Wu Member, IEEE,Jin A. Kong Fellow, IEEE, and Daniel W. Engels Member, IEEE

I

Proceedings of the 28th IEEEEMBS Annual International ConferenceNew York City, USA, Aug 30-Sept 3, 2006

SaA05.5

1-4244-0033-3/06/$20.00 ©2006 IEEE. 4340

In this paper we describe a method and apparatus for

accurately maintaining sample temperature, while the

sample is being exposed to RFR, under conditions, where

in the absence of such control, the sample temperature

would increase substantially. We show that this

methodology and assembly is capable of maintaining the

sample temperature within + 0.5oC precision, irrespective

of the RFR frequency, intensity, or initial sample

temperature. This methodology provides an accurate way

to investigate non-thermal effects of RFR on a variety of

samples.

II. MATERIALS AND METHODOLOGY

A. General

The test apparatus was composed of five physical sub-

systems: the test sample subsystem, the radiation exposure

(RF exposure) subsystem, the power measurement

subsystem, the temperature measurement and control

subsystem, and an overall control subsystem. The

equipment required for the apparatus is listed in Table 1;

Figure 1 provides an equipment connection diagram.

In the following subsections, we describe each of these

subsystems.

Fig. 1. Equipment connection diagram. Equipment labels shown are

model numbers for devices listed in Table 1. (See Table 1)

B. Test sample subsystem

The test chamber was a TEM cell - a wave guide that

allows propagation of electromagnetic plane waves in the

TEM mode, also called the fundamental or TM0 mode [7].

As shown in Figure 2, the cell (Montena TEM-F3000) was

fabricated with a copper plate suspended in the middle of

the chamber. This plate was used as a platform for

supporting the sample – 1.5mL of aqueous solution in a

standard 1.8mL vial – in a hand-made supporting “bed” of

Rohacell, a material which is transparent to radiation at

radio and microwave frequencies. The bed ensured that

the vial made physical contact with the upper inner surface

of the TEM cell and was fixed in place by friction (Figure

3). The septum of each vial was punctured to create a hole

about the same size as the temperature sensor (Ipitek

Lumitherm LT-X5RS). The sensor was first inserted

through a hole in the door of the TEM cell and then

through a tiny funnel into the vial. This technique

minimized evaporative and leakage losses and sensor

damage.

TABLE 1

EQUIPMENT SPECIFICATION

Make and model Description

Agilent 82357A USB/GPIB interface for Windows

Cable X-perts

CXP1318FN

50 Ohm, low loss, 3ft coaxial cable with “N”

male connectors

Dell PC Computer running MATLAB 7

Hewlett Packard 438A Dual input power meter with remote capabilities

and 5-ft sensor cables

Hewlett Packard 437B Single input power meter with remote

capabilities and 5 ft sensor cables

Hewlett Packard

8482A

100mW coaxial power sensor; range: 1μW to

100mW

Hewlett Packard

8482B

25W coaxial power sensor with power

dissipation and heat sink; range: 1W to 25W

Ipitek Lumitherm LT-

X5R

Fiber optic thermometer processing unit with

remote capabilities

Ipitek Lumitherm LT-

X5RS

Fiber optic temperature sensor

MECA 722N-30-

1.650W

Dual directional coupler; range: 800MHz to

2.5GHz

Montena TEM-F3000 Closed TEM cell for measurements up to 3GHz;

50 Ohm load required

New Brunswick

Scientific Comparny

G24 environmental incubator shaker (not show

in Figure 1)

Ophir 5151R Radio frequency power amplifier, 0.8 – 2.5

GHz, 25Watt

Rohde and Schwarz

SML03

9kHz to 3.3GHz signal generator with remote

capabilities

TE Technology CP-

031

Thermo-electric cooler, -20°C to 100°C. Also

known as a peltier cooler.

TE Technology TC-

24-25 RS232

Thermo-electric temperature controller and

power supply with remote capabilities

TE Technology TP-1 Thermal paste (not show in Figure 1)

Miscellaneous items GPIB cables, network cable (RJ45), 12AWG

wire (cooler interconnection), serial cable, etc.

With the exception of the miscellaneous cables, all devices listed above

are shown in the connection diagram (Fig. 1) using model numbers only.

C. RF exposure subsystem

The signal generator (Rohde & Schwarz SML03),

power amplifier (Ophir 5151R), dual directional coupler

(MECA 722B-30-1.650W) and TEM cell were connected

in series using coaxial cables (Cable X-perts

CXP1318FN), satisfying the 50 input/output impedance

requirements for several of these devices. Two of the four

ports on the coupler and one of the two ports on the TEM

cell were used for interconnection while the remaining

three ports were used for power measurements.

4341

Fig. 2. Partial section through TEM cell, showing vial and supporting

“bed”

Fig. 3. Rohacell “bed” for supporting sample vials within the TEM cell.

D. Power measurement subsystem

The directional coupler provided a power attenuation of

approximately 30dBm between the input and the coupled

ports, one sampling the incident power from the power

amplifier and the other, the reflected power from the TEM

cell. Sensors (Hewlett Packard (HP) 8482A) were

connected to each of the coupled ports and to channels A

and B of the dual input power meter (HP 438A). A third

sensor (HP 8482B) was connected to the output of the

TEM call and to the single channel power meter (HP

437B) in order to measure the transmitted power.

Due to input power constraints for the sensors (Table 1)

and gain/attenuation provided by the amplifier and

coupler, the control software was designed to avoid

damage to physical components by limiting user inputs.

E. Temperature measurement and control subsystem

The test sample subsystem was placed in a controlled

temperature oven (New Brunswick Scientific Company

G24 environmental incubator shaker) which allowed

manual control of the ambient temperature within the

TEM cell. Sample temperature was determined using the

fiber optic thermometer (Ipitek Lumitherm LT-X5R): the

sensor was connected to port 1 of the thermometer and its

tip immersed in the sample, as described in part B.

With the sample vial in contact with the upper inner

surface of the TEM cell as indicated in partB, the cooler

was positioned above the sample but on the upper external

surface of the cell. Thermal paste (TE Technology TP-1)

applied to the surface of the cooler improved thermal

conductivity. This arrangement maximized heat exchange

between the cooler and the sample.

F. Control subsystem

Computer controllable devices were connected to the

Dell PC by one of three methods. First, the signal

generator and power meters were connected via GPIB

cables to the GPIB interface (Agilent 82357A), which was

plugged into one of the USB ports on the computer.

Second, the thermometer was directly connected via

network cable and used TCP/IP for communication. Third,

the cooler controller was plugged into the computer’s

COM1 port via serial cable.

All the devices were manipulated from scripts

(programs) in the MATLAB 7 environment, which

contained the visa-gpib, serial and tcpip instrument

functions necessary for communication. The final program

allowed the user to choose one of four exposure

conditions: temperature control only, RFR exposure only,

both conditions simultaneously or neither. User inputs

included incident RFR frequency and power intensity, the

required temperature and the exposure duration. The script

stored all data in text-based LOG (.log) files and provided

a processing option to plot data and perform specific

absorption rate (SAR) calculations.

III. EXPERIMENTAL RESULTS AND DISCUSSION

Herein we have developed a procedure and a set-up to

accurately and precisely control and maintain temperature

of a sample solution when irradiated with RFR. Evaluation

of the implemented system required two sets of

experiments, the first to show the effects of RFR on

sample temperature, and the second to determine whether

the system designed could limit such effects. Two

different frequencies, 2.45 GHz and 915 MHz were used

to evaluate their effects on sample temperature at two

different incident radiation power levels of 22W and 4W.

Figure 4 shows the effect of RFR frequency, power

intensity and initial sample temperature on time-course of

sample temperature in the apparatus described in Part II,

and without direct sample cooling. Clearly, an increase in

sample temperature was observed in all cases, except for

in the case of 915 MHz at 4W incident power and 30oC.

The extent of the increase was dependent on the RFR

frequency, RFR intensity and initial sample temperature.

The maximum increase in sample temperature, T 9oC,

was observed in the case of 2.45 GHz RFR at 22W

incident radiation power and initial sample temperature

4342

50oC. The extent and rate of sample temperature increase

is reduced by using a lower frequency, power and initial

sample temperature.

Fig. 4. Effect of the frequency, intensity, and initial sample temperature

on the time-course of sample temperature upon exposure to

radiofrequency radiation. 2.45 GHz at 22W (open squares), 915 MHz at

22W (open circles), 2.45GHz at 4W (crosses) and 915 MHz at 4W (open

triangles). The curve obtained with 2.45 GHz at 4W at initial sample

temperature of 50oC overlays that of 915 MHz at 22W at the same initial

temperature.

To determine whether the apparatus met the stated goal of

maintaining sample temperature regardless of RFR

exposure, the performance of the apparatus was evaluated

under the harshest conditions investigated in the first set of

experiments – 2.45GHz, 22W incident power and 30°C

initial temperature.

Fig. 5. Effect of the external peltier cooling device on sample

temperature upon exposure to radiofrequency radiation at 2.45 GHz and

22W incident power. No external cooling (open circles); with external

cooling (Open squares).

Figure 5 compares the time-course of sample

temperature as the sample was exposed to the stated

conditions without active cooling to that with active

cooling. With cooling, the sample temperature stabilized

at an average of 50.5oC, with a maximum variation of

+0.4oC for up to 24 hours.

IV. CONCLUSIONS

This study demonstrates that radio-frequency radiation

at 915MHz and 2.45GHz increased sample temperatures

but that the assembled system reported here significantly

limited RFR induced-heating in aqueous samples. Use of

this equipment and its software allows tests with any

given RFR/temperature combination. Therefore, the

apparatus allows to investigate RFR-induced non-thermal

effects (physical/chemical) in aqueous solutions of such

pharmaceuticals as drugs and biologics at any RFR

frequency, power intensity and sample temperature

combination.

ACKNOWLEDGMENT

The authors thank Robin Koh and Tom Scharfeld for

their early work in establishing this project. The authors

also thank Dennis Kim, Jim Dowden and Tom Pizzuto for

their support and encouragement of this work.

REFERENCES

[1] U.S. Department of Health and Human Services, Food and Drug

Administration, Combating Counterfeit Drugs: A Report of the

Food and Drug Administration, February 18, 2004. Available:

http://www.fda.gov/oc/initiatives/counterfeit/report02_04.html

[2] J. L. Kirschvink, “Microwave absorption by magnetite: a possible

mechanism for coupling nonthermal levels of radiation to

biological systems,” Bioelectromagnetics, vol. 17, pp. 187-194,

1996.

[3] E. Marani and H. K. P. Feiraband, “Futire perspectives in

microwave applications in life sciences,” Eur. J. Morphol., vol. 32,

pp. 330-334, 1994.

[4] M. Porcelli, G. Cacciapuoti, S. Fusco, R. Massa, G. d`Ambrosio, C.

Bertoldo, M. D. Rosa, V. Zappia, “Non-thermal effects of

microwaves on proteins: thermophilic enzymes as model system,”

FEBS Lett., vol. 402, pp. 102-106, 1997.

[5] D. I. de Pomerai, B. Smith, A. Dawe, K. North, T. Smith, D. B.

Archer, I. R. Duce, D. Jones, and E. P. M. Candido, “ Microwave

radiation can alter protein conformation without bulk heating,”

FEBS Lett., vol. 543, pp. 93-97, 2003.

[6] E. Bismuto, F. Mancinelli, G. d’Ambrosio, and R. Massa, “Are the

conformational dynamics and the ligand binding propertoes of

myoglobin affected by exposure to microwave radiation?,” Eur.

Biophys. J., vol. 32, pp. 628-634, 2003.

[7] J. A. Kong, Electromagnetic Wave Theory, EMW Publishing,

Cambridge MA, USA.

25

30

35

40

45

50

55

60

65

0 10 20 30 40 50 60

Time, min

Temperature,oC

25

30

35

40

45

50

55

60

65

0 10 20 30 40 50 60

50.5 + 0.4 (24 hr)

Time, min

Temperature,oC

4343

ANEXO 2

Report to FDA of PQRI RFID Working Group

by Robert H. Seevers, Ph.D.

December 1, 2005

RFID Report to FDA Page 2 of 14

Report to FDA from the PQRI RFID Working Group Executive Summary In its report, Combating Counterfeit Drugs, issued in February 2004 (1), the FDA stated that “Radiofrequency Identification (RFID) tagging of products by manufacturers, wholesalers, and retailers appears to be the most promising approach to reliable product tracking and tracing.” The Agency, while clearly indicating in the same report (1) that none of the contemplated track and trace technologies, including RFID, represents a “magic bullet”, has encouraged regulated industry to consider RFID as a potential tool in anti-counterfeiting efforts. A concern has been raised about the potential impact on product quality due to exposure of drug products to radiofrequency radiation which would occur if RFID is adopted. This issue was assigned to the Drug Product Technical Committee (DPTC) of PQRI. The DPTC requested a study design for exposure of biological/biotech drug products to RF radiation; in the course of discussing the proposed study, the DPTC unanimously requested that a risk assessment for such exposure be done. This report is that assessment. There are three main components to the risk assessment performed: 1) a summary of the results from the industry sponsored, FDA reviewed Accenture Jumpstart test of the impact of RF exposure on product quality and product temperature specifically (2,3), 2) a review of the scientific literature on the impact of RF exposure on biomolecules, and 3) a discussion of the perspective on risk provided by high-field NMR studies of biomolecules. The key finding of the Accenture Jumpstart effort in regard to any potential product quality impact from exposure to up to 16 hours of RF is a temperature rise of between one half and one degree Celsius and no change in strength, purity, and potency (2,3). Given that the ICH stability studies allow a ±2ºC variability in chamber temperature, such a temperature rise should be considered negligible. A review of the current literature on RF exposure of biomolecules indicates that the effects noted in both the theoretical and experimental papers (5,6) are understandable as extremely localized thermal effects. To extrapolate these effects to the situation of RF exposure through the use of RFID track and trace technology would require a drug product that is sufficiently thermally labile and concentrated so that these potential highly localized effects could propagate through the sample to a sufficient extent to create a measurable impact. Under the conditions anticipated in the drug distribution environment, this is most unlikely. A recent paper from FDA’s Center for Devices and Radiological Health (CDRH) is a simulation of RF exposure under theoretical worst case scenarios (7). Although the abstract and the paper state that the modeled exposure is for insulin, a careful reading of the paper reveals that insulin was modeled by a gel of 0.9% saline with

RFID Report to FDA Page 3 of 14

hydroxyethylcellulose. This study produced results that are entirely consistent with the Accenture Jumpstart study (2,3); that is, exposure to RF under a worst case scenario will not increase the temperature of a pharmaceutical by more than 2°C, again a negligible change. The key to an in-depth risk assessment of drug products, in particular, biomolecules, exposed to RF due to the use of RFID is the realization that the extremely high magnetic fields used in current NMR studies of biomolecules require similarly high RF input in the 800-900 mHz range, precisely the range anticipated for most US RFID systems. This means that NMR structural studies of biomolecules provide an appropriate comparator to RFID for the purpose of determination of relative risks. The field strength of the NMR is more than a million fold greater than that experienced by a biomolecule exposed to RF from an RFID tag. The drug will typically also be exposed to this miniscule field for less than one millionth the time that biomolecules are exposed to RF in NMR structural determination experiments. That exposure to RF in this energy range does not cause non-thermal damage to biomolecules is born out by a number of recent studies that used NMR to examine the active site(s) of several proteins, including the binding of ligands to those proteins (10-15). In these peer-reviewed papers, NMR studies on several different proteins are presented, including: human acetylcholinesterase (10), small ubiquitin-related modifier (SUMO) proteins (11,15), human factor Xa protein (12), and heme oxygenase of both human (13) and bacterial origin (14). The adoption of a radiofrequency identification track and trace system may contribute significantly to public health by making it harder for counterfeiters to slip into the legitimate distribution system. There have been concerns raised regarding the possibility of damage to drugs, particularly biological/protein based drugs, due to exposure to the RF energy that such a system would entail. This paper examines the risks involved and characterizes them as thermal or non-thermal. The thermal risks have been determined to be negligible by the Accenture Jumpstart study which showed only a very small temperature rise on exposure to several hours of RF from an RFID system. The non-thermal risks are assessed by examining representative articles from the peer-reviewed literature and by considering the NMR structural studies of biologicals and proteins which employs RF in the same energy range as RFID. It is seen that what the current literature presents as so-called non-thermal effects are, in fact, highly localized thermal effects. Concerns over such effects are unwarranted, as they are improbable for any biological or protein-based drug that is stable at either controlled room temperature or 2-8ºC. Further, a recent paper from FDA’s Center for Devices and Radiological Health (7) simulated exposure of liquid drugs in vials to RF under worst case conditions and observed a less than 2ºC rise in temperature, consistent with the Accenture Jumpstart results. It was further seen that the relative time risk of exposure to RF due to RFID is greater than 2.5 million times higher for NMR than for RFID, except in the artificial worst case

RFID Report to FDA Page 4 of 14

scenario of a package stuck for 16 hours in an RFID reader that remains on. Consideration of the relative distance involved in NMR vs. RFID led to a relative risk greater than 1 million times NMR. And yet, as noted above, NMR can and has been used not simply for protein structural determination, but to actually measure binding at the active site of the protein, in other words, as a bioassay. All of the above, taken together, provide ample evidence that exposure to RF energy due to the implementation of an RFID track and trace system represents a negligible risk to drug product quality, whether of small molecules or biological or protein-based drugs.

RFID Report to FDA Page 5 of 14

Report to FDA of PQRI RFID Working Group Background & Problem Statement In its report, Combating Counterfeit Drugs, issued in February 2004 (1), the FDA stated,

Because the capabilities of counterfeiters continue to evolve rapidly, there is no single "magic bullet" technology that provides any long-term assurance of drug security. However, a combination of rapidly improving "track and trace" technologies and product authentication technologies should provide a much greater level of security for drug products in the years ahead. Similar anti-counterfeiting technologies are being used in other industries, and FDA intends to facilitate their rapid development and use to keep drugs secure against counterfeits.

Specifically, in regard to RFID, the same report indicated that:

Radiofrequency Identification (RFID) tagging of products by manufacturers, wholesalers, and retailers appears to be the most promising approach to reliable product tracking and tracing. Significant feasibility studies and technology improvements are underway to confirm that RFID will provide cost-reducing benefits in areas such as inventory control, while also providing the ability to track and trace the movement of every package of drugs from production to dispensing. Most importantly, reliable RFID technology will make the copying of medications either extremely difficult or unprofitable. FDA is working with RFID product developers, sponsors, and participants of RFID feasibility studies to ensure that FDA's regulations facilitate the development and safe and secure use of this technology. FDA is also working with other governmental agencies to coordinate activities in this area.

Thus, the Agency, while clearly indicating in the same report (1) that none of the contemplated track and trace technologies, including RFID, represents a “magic bullet”, has encouraged regulated industry to consider RFID as a potential tool in anti-counterfeiting efforts. As a follow-up the Agency released a Compliance Policy Guide (CPG), Section 400.210 Radiofrequency Identification Feasibility Studies and Pilot Programs for Drugs in November of 2004 (4). This CPG stated:

To the extent that it may be necessary, FDA intends to exercise enforcement discretion as described below for studies that fall within all of the following parameters: • A manufacturer, repackager, relabeler, distributor, retailer, or others acting at

their direction will attach RFID tags (chips and antennae) to only immediate containers, secondary packaging, shipping containers, and/or pallets of drugs that are being placed into commerce. There is no limit to the number of tags or readers that may be used in the study.

• The drugs involved will be limited to prescription or over-the-counter finished products. The drugs involved will not include those approved under a Biologics

RFID Report to FDA Page 6 of 14

License Application or protein drugs covered by a New Drug Application. The study need not have a pre-determined time limit or endpoint, except that tag placement for the study will be completed by December 31, 2007.

• … • The tag readers will work by emitting electromagnetic energy at radio frequencies

of 13.56 megahertz, 902-928 megahertz, or 2.4 gigahertz, and at powers in compliance with regulatory requirements of the Federal Communications Commission (i.e., 1-4 watts effective isotropically radiated power).

In summary then, the FDA has actively promoted RFID use as one component of an overall strategy to secure product integrity and further protect patient safety. One key question was whether exposure to radiofrequency energy in the range contemplated would be expected to have any impact on the quality of the drug products so tagged. Particular concern was expressed by the Agency in regard to Biologics or protein drugs as noted above. The purpose of the PQRI RFID taskforce was to evaluate to the risk to product quality posed by RF exposure of Biologics and protein based drugs. The DPTC requested a study design for exposure of biological/biotech drug products to RF radiation; in the course of discussing the proposed study, the DPTC unanimously requested that a risk assessment for such exposure be done. This report is that assessment. Accenture Jumpstart Results Project Jumpstart was formed in February 2004 by Accenture, a leading business consulting and services firm, to promote pilots of RFID technology in the pharmaceutical industry supply chain. It included consumer goods and pharmaceutical manufacturers, distributors, retailers, and associations. In September 2004, some Project Jumpstart members completed a 12-month EPC pilot project that was unprecedented in its scope and pharmaceutical industry participation (2,3). The participants spent months designing the pilot then conducted an eight-week trial that tracked 10 products from multiple manufacturers through 16 business scenarios at 15 locations. More than 13,000 products were tagged and read. HDMA participated in the project, as did Abbott Laboratories, Barr Laboratories, Cardinal Health, CVS Pharmacy, Johnson & Johnson, McKesson, National Association of Chain Drug Stores (NACDS), Pfizer, and Rite Aid, with Accenture serving as project administrator. The project was designed as a proof of concept for using EPC technology in multiple business operations. The main focus was to determine how unit-level item serialization could improve supply chain security. Other goals included:

• Assess RFID suitability to satisfy regulatory mandates such as the Florida pedigree requirement;

• Establish processes to facilitate returns and recalls; • Develop and execute tests to see if electromagnetic energy from radio frequency

affects the efficacy, potency, and strength of solid drugs included in the trial;

RFID Report to FDA Page 7 of 14

• Gain knowledge and develop solutions for labeling, including tag frequency and the color, size, wording, and location of RFID labels.

Jumpstart declared the project a success and reported it demonstrated the potential of RFID technology to provide many benefits specific to the industry, including meeting regulatory transaction history requirements, returns and recall management, improved consumer safety through electronic track-and-trace and authentication capabilities, greater order accuracy, and increased labor productivity. Several obstacles and challenges to widescale implementation also were identified. The project provided many insights and recommendations as to how companies can develop RFID strategies and implement the technology, and how the industry needs to collaborate to further develop EPC systems to meet specific needs. In particular, the potential impact of RF exposure on product quality was evaluated. The FDA requested that the manufacturers who participated in EPC Jumpstart test a representative sample of the products they intended to tag by exposing them to electromagnetic energy. The FDA requested that these tests should be executed prior to introducing tagged product into commercial distribution. Furthermore, it was requested that the tests be designed so that the effect of the exposure to the electromagnetic energy on the strength, purity, and potency of the exposed product be compared to the strength, purity, and potency of unexposed control product. The FDA indicated that they were primarily concerned about the thermal effects of exposure to the electromagnetic fields emitted by RFID systems. The FDA suggested the following parameters while creating the test protocol: 1. A continuous exposure of the product to electromagnetic energy for the longest

period of time that a product would remain in the maximum field strength produced by the RFID antenna. The duration of 16 hours is considered to provide a worst-case scenario simulating a package containing pharmaceutical product being left overnight in the RFID reader field.

2. The product should be placed at the closest possible distance to the antenna and exposure of the product to electromagnetic energy should use equipment and other environmental factors that simulate the worst-case heating conditions expected during commercial use. In considering the 60 degree beam width for the antenna, a distance of 12 inches provided an exposure field approximately 12 inches in diameter.

Using these requested parameters by the FDA and creating additional parameters within the Pharma EPC Jumpstart Protocol the EPC Jumpstart Project exposed each product to a controlled duration of fixed radio frequency energy using 915 MHz Stationary Reader/Antenna Set. This was considered a worst-case in that it provided a maximum gain and power, operating at a full 4 watts power output. The participating manufacturers exposed their EPC Jumpstart Products (and some additional products including a liquid) to electromagnetic energy. The following conditions existed and equipment was used to capture the effects:

RFID Report to FDA Page 8 of 14

1. The product was exposed to the RF energy in the pharmacy shelf package

configuration with the RFID tag affixed to the package (Package information per product tested within EPC Jumpstart is attached).

2. A field strength meter (Agilent 84110EM) was used to confirm that the 915 MHz reader/antenna was performing as expected and was operating consistently across the area where the product was located. This information was recorded periodically through the test along with the ambient field strength.

3. Temperature of the interior of the container with product present was monitored to evaluate temperature observed during the exposure period. A Teflon temperature probe was placed in each package during the exposure to RF energy as well a temperature probe measuring ambient room temperature. Continuous temperature data was captured using a 16 channel evaluation system.

4. Three replicates of each product were exposed to the RF Field. Once the exposure was complete, the products were returned to the manufacturers to be tested using currently approved analytical methods. Typical stability indicating methods were employed (e.g., physical characteristics, dissolution, assay, degradation, etc.). In cases where the routine release testing and stability testing for a particular test parameter are different, the stability method was used, either in conjunction with or in replacement of the routine release method. Where multiple packages of product were exposed, testing was performed on individual samples taken from each package. All manufacturing companies have reviewed the results, and have determined that there are no apparent effects on strength, purity, and potency of exposed product to electromagnetic energy when compared to the controls. The results have shown the following: 1. Approximately a one-half degree increase in temperature throughout the 16 hour

period for solids. Approximately a one degree increase in temperature was noted throughout the 16 hour period for the one liquid product tested.

2. Field Strength remained constant throughout the 16 hour period. 3. Strength, purity, and potency were not affected by the exposure. While none of the products evaluated in the Jumpstart project were biologicals or protein drugs, it can be concluded from this study that thermal effects of RFID exposure on any drug whether small molecule or biological are not expected to pose any risk to product quality. A copy of the Jumpstart report was sent to Dr. Paul Ruldolph, then at FDA (2,3). RF Exposure Papers discussion In recent years the question of whether exposure to radiofrequency energy could have an effect on biological materials has been raised. While some of what has been put forth can be dismissed as anecdotal at best, there have been peer-reviewed scientific

RFID Report to FDA Page 9 of 14

articles that address this question. It is appropriate to evaluate what is known in this area as the potential impact of RF exposure on the quality of biological and protein based drugs due to the use of RFID technology is considered. While it is not possible to review the entire literature on this subject, three representative papers will be considered, one theoretical, one experimental, and the third a simulation of RF exposure of insulin vials. In none of the papers is a convincing scenario provided that RF exposure of the type expected for the use of RFID track and trace technology, even a worst case situation, presents a credible risk to product quality. The theoretical paper is by an Australian group and was published in 2000 (5). The authors state that microwave exposure under “athermal” conditions occurs when no temperature rise can be measured by conventional thermometry. They note that existence of biological effects arising from such athermal exposure is still controversial, significantly because of a lack of the linear dose response relation. They propose a model in which pulsed microwave radiation causes a triggering of the heat shock or stress response by altering the conformation of proteins through a transient heating of the protein and its close environment. In support of this model, the authors use the heat diffusion equation and show that pulsed exposure even when athermal can lead to transient temperature excursions outside the normal range. They thus propose that the power window phenomenon in which biological effects are observed at low power levels may be caused by an incomplete triggering of the heat shock response. The authors focus on the heat shock response of cells as the mechanism by which non-thermal RF exposure might produce biological effects. This is presumably because the heat shock proteins are exquisitely sensitive to energy input and readily undergo conformation changes. The authors offer the following scenario:

At low power levels, a partial unfolding of specific target protein(s) occurs, which will be insufficient to induce the stress response, but sufficient to alter protein function. A biological effect (e.g. on cell proliferation) will be observed. At higher power levels a more unfolded (molten globule) conformation is induced. The stress response will be activated, protecting the protein, and preventing an observable biological effect. At very high power levels, protein aggregation and precipitation occurs, and despite the activation of the entire stress response, a catastrophic biological effect (e.g. cell death) will be observed.

Thus the theoretical effects the authors consider in a non-thermal situation are, in fact, thermal effects that occur in a highly localized environment on a scale and time frame too small to be observed by measurements of the temperature of a whole sample. In addition, the postulated effects are primarily whole-cell focused and where they do center on specific proteins, only those proteins that are particularly sensitive to thermal effects are likely to be impacted.

RFID Report to FDA Page 10 of 14

In terms of the anticipated risk to biological or protein based drugs from exposure to RF, whole cell effects can generally be discounted as most materials in commerce are either isolated, purified molecules or dead cells in the case of vaccines. Live cell vaccines are theoretically at some risk under this scenario, but the amount of that risk is too small to quantify. The risk to molecules of highly localized thermal effects and unfolding can be evaluated by reference to bulk stability data for the material. The only molecules that one would anticipate to be at risk in this situation are those that are so thermally labile that they could never be formulated or marketed. That is, any localized effect would create an impurity in such small amounts as to be undetectable by any assay. The second paper attempts to establish experimentally what the first offered in theory: that RF exposure can change protein conformation without bulk heating (6). The authors exposed concentrated solutions of bovine serum albumin (BSA) at various temperatures to microwave radiation and observed protein aggregation as determined by light scattering. In a separate experiment, they observed that a heat shock gene recombinantly expressed in E coli was expressed more in microwave exposed cells than in controls. The concentration of BSA used in the experiments was reported to be 50-100 mg/mL. For perspective, the concentration of insulin in marketed preparations is in the 1-5 mg/mL range. Thus, the effect observed is (intentionally) magnified by the use of a very highly concentrated solution. Also, the authors controlled temperature through the use of incubators and by measuring with a microthermocouple after exposure. What this experimental set up did not do is provide for or document adequate thermal control of the samples intended to be kept at a specific temperature. As will be seen below in the NMR section, RF exposure can and does have a significant local heating effect. It is likely that the effects observed in this study are attributable to local heating rather than a non-thermal effect. The other significant consideration in terms of risk evaluation is the power used. The microwave radiation employed was 1.0 GHz and 0.5 W. While the geometry of the experimental set up was not precisely described in the paper, it is clear that the proximity to the RF source was likely considerably closer than is anticipated in the RFID situation. The third paper is from FDA’s Center for Devices and Radiological Health (CDRH). It presents a simulation of RF exposure under theoretical worst case scenarios (7). Although the abstract and the paper state that the modeled exposure is for insulin, a careful reading of the paper reveals that insulin was modeled by a gel of 0.9% saline with hydroxyethylcellulose. The purpose of the gel was to minimize or eliminate diffusion effects on any temperature rise observed. The modeling conditions used 200 - 350 Watts of power rather than the 1 - 4 Watts typical of RFID devices. This was to enable determination of temperature effects over very short time frames to minimize or eliminate diffusion effects. These results were then extrapolated to the real world situation. The results extrapolate to a 1.7°C temperature rise in the single case of a vial sitting in the position of maximum effect. In any other position, or if the vial were surrounded by other vials as would be the case in a commercial package, the effect was

RFID Report to FDA Page 11 of 14

estimated to be 0.6°C or less. The author's extrapolation applies to a worst case scenario that neglects heat lost to the surrounding environment. In short, this study produced results that are entirely consistent with the Accenture Jumpstart study (2,3); that is, exposure to RF under a worst case scenario will not increase the temperature of a pharmaceutical by more than 2°C. In summary, the effects noted in the first two papers (5,6) are understandable as extremely localized thermal effects, while the third paper (7) presents simulated RF exposure under a worst case scenario that results in a temperature rise of less than the ±2ºC variability permissible in stability chamber temperature; such a temperature rise should be considered negligible. . To extrapolate these effects to the situation of RF exposure through the use of RFID track and trace technology would require a drug product that is sufficiently thermally labile and concentrated so that these potential highly localized effects could propagate through the sample to a sufficient extent to create a measurable impact. Under the conditions anticipated in the drug distribution environment, this is most unlikely. Similarity of RF Exposure in RFID Use to NMR Studies In evaluating the relative risk to product quality of biologics and protein based drugs exposed to RF energy, a comparison with nuclear magnetic resonance (NMR) studies of biomolecules structure is helpful. NMR has been used to determine the structures of small molecules for half a century and has been applied to larger molecules for the past two decades. The key point to consider here is the Larmor Equation which states that, in the presence of an external magnetic field, the energy of interaction of the proton’s own magnetic field is proportional to the strength of the external magnetic field. The extremely high magnetic fields used in current NMR studies of biomolecules require similarly high RF input in the 800-900 mHz range, precisely the range anticipated for most US RFID systems. This means that NMR structural studies of biomolecules provide an appropriate comparator to RFID for the purpose of determination of relative risks. The following table lays out a comparison of the different factors involved in both high field NMR and RFID (8,9).

Parameter High Field Protein NMR RFID RF Frequency (8,9) 600-900 MHz 13.56 MHz or 902-928 MHz Power (at antenna) (8) 5-30 W 4 W Exposure Time (8) Several days with ~10-25%

duty cycle msec; worst case assumed to be 16 h

Typical Protein Concentration (8)

As close to solubility limit as possible, typically millimolar

Insulin for example, is present in vials at ~ 4 mg/mL

Distance from RF antenna (8) 2-3 mm 2-3 meters Calculated Field Strength (9) >2.4 MA/m < 1A/m

RFID Report to FDA Page 12 of 14

Let us calculate the relative risk factors from the information above. Assume an NMR exposure of 3 days with a 10% duty cycle and a 10 millisecond RFID exposure. The relative time of exposure is 2,592,000 greater for the NMR scenario. Only in the artificial assumed worst case of a 16 hour RFID exposure, would the relative time of exposure become roughly equivalent at 0.45 in favor of the RFID exposure. A second risk factor is the distance from the antenna. The typical NMR setup has the RF coils as close as possible to the sample. The typical RFID scanner is designed to be used to detect goods passing through a loading dock. The difference is then a factor of a 1000 (millimeters vs. meters). Since the field strength will be determined by the square of the distance this represents a minimum of a relative risk factor of 1,000,000 greater for NMR. Note that a separate calculation in reference 9 produces relative risk factor in terms of field strength that is 2,400,000. In summary, the field strength of the NMR is more than a million fold greater than that experienced by a biomolecule exposed to RF from an RFID tag. The drug will typically also be exposed to this miniscule field for less than one millionth the time that biomolecules are exposed to RF in NMR structural determination experiments. NMR Studies and Bioassay That exposure to RF in this energy range does not cause non-thermal damage to biomolecules is born out by a number of recent studies that used NMR to examine the active site(s) of several proteins, including the binding of ligands to those proteins (10-15). In these peer-reviewed papers, NMR studies on several different proteins are presented, including: human acetylcholinesterase (10), small ubiquitin-related modifier (SUMO) proteins (11,15), human factor Xa protein (12), and heme oxygenase of both human (13) and bacterial origin (14). What the referenced studies have in common is that they each use high field NMR, with the correspondingly high RF input to study the three-dimensional active structure of the respective proteins. As part of each study involved using NMR techniques to determine binding to the active site of the protein being studied, these reports can be characterized as NMR-based bioassays. In short, exposure to the RF in the same range as is anticipated for an RFID system not only did not cause any detectable non-thermal damage to the proteins involved; that exposure was an integral part in performing a bioassay of those proteins. Further, the work cited includes studies that demonstrate the NMR structure determination provides results fully consistent with crystal structure data (14) and have documented the presence of individual hydrogen bonds at the active site of human acetycholinesterase (10). That last observation is critical; if the hydrogen bonds, weaker than any covalent bond, can be observed in these studies while a highly concentrated solution of the protein is undergoing exposure to high energy RF radiation, it clearly demonstrates that such exposure does not disrupt these interactions. Conclusion

RFID Report to FDA Page 13 of 14

The adoption of a radiofrequency identification track and trace system may contribute significantly to public health by making it harder for counterfeiters to slip into the legitimate distribution system (1). There have been concerns raised regarding the possibility of damage to drugs, particularly biological/protein based drugs, due to exposure to the RF energy that such a system would entail. This paper has examined the risks involved and characterized them as thermal or non-thermal. The thermal risks have been determined to be negligible by the Accenture Jumpstart study which showed only a very small temperature rise on exposure to several hours of RF from an RFID system (2,3). The non-thermal risks have been assessed by examining representative articles from the peer-reviewed literature and by considering the NMR structural studies of biologicals and proteins which employ RF in the same energy range as RFID. It was seen that what the current literature presents as so-called non-thermal effects are, in fact, highly localized thermal effects. Such concerns are improbable for any biological or protein-based drug that is stable at either controlled room temperature or 2-8ºC. It was further seen that the relative time of exposure to RF due to RFID is greater than 2.5 million times higher for the typical NMR experiment versus the normal RFID exposure. Only in the artificial worst case scenario of a package stuck for 16 hours in an RFID reader that remains on was the exposure time comparable to the NMR protein experiment. Consideration of the relative distance involved in NMR vs. RFID led to a relative risk factor for field strength greater than 1 million times for NMR over RFID exposure. And yet, as noted above, NMR can and has been used not simply for protein structural determination, but to actually measure binding at the active site of the protein, in other words, as a bioassay. All of the above, taken together, provide ample evidence that exposure to RF energy due to the implementation of an RFID track and trace system represents a negligible risk to drug product quality, whether of small molecules or biological or protein-based drugs. References 1. Combating Counterfeit Drugs: A Report of the Food and Drug Administration,

February 2004 (link). 2. Letter from Jamie T. Hintlian, Partner, Accenture LLP, to Dr. Paul Rudolph of FDA,

dated July 6, 2004. 3. RF Testing Summary Slides, Accenture LLP, 2004. 4. Compliance Policy Guide, Section 400.210 Radiofrequency Identification Feasibility

Studies and Pilot Programs for Drugs in November 2004 (link).

RFID Report to FDA Page 14 of 14

5. Biological Effects of Electromagnetic Fields—Mechanisms for the Effects of Pulsed Microwave Radiation on Protein Conformation, JA Laurence, PW French, RA Lindner, and DR Mckenzie, J. theor. Biol. (2000) 206, 291-298.

6. Microwave Radiation Can Alter Protein Conformation Without Bulk Heating, DI de Pomerai, B Smith, A Dawe, K North, T Smith, DB Archer, IR Duce, D Jones, EPM Candido, FEBS Letters 543 (2003) 93-97.

7. Liquid Pharmaceuticals and 915 MHz Radiofrequency Identification Systems, Worst-Case Heating and Induced Electric Fields, H Bassen, RFID Journal, 2005 (link).

8. Telephone conversation with Dr. Wesley Cosand of Bristol Myers Squibb, a biological NMR expert, Patrick Sweeney II, CEO of Odin Technologies, an RFID expert, and Robert Seevers, author of this document, May 25, 2005.

9. Assessment Of The Potential Radiation Effects Of Aegate RFID Scanner On Drugs, T. Fergus, S. Crammond, and N. Butt. PA Consulting Group 2005. Note that copies of this reference have been provided to Jon Clark, Guirag Poochikian, and Raj Upoor of the FDA.

10. Short, Strong Hydrogen Bonds at the Active Site of Human Acetylcholinesterase: Proton NMR Studies, M.A. Massiah, C. Viragh, P.M. Reddy, I.M. Kovach, J. Johnson, T.L. Rosenberry, and A.S. Mildvan, Biochemistry 2001, 40, 5682-5690.

11. Identification of a Substrate Recognition Site on Ubc9, D. Lin, M.H. Tatham, B. Yu, S. Kim, R.T. Hay, and Yuan Chen J Biol Chem Vol. 277, No. 24, 21740–21748, 2002.

12. Exploring the Active Site of Human Factor Xa Protein by NMR Screening Small Molecule Probes, L. Fielding, D. Fletcher, S. Rutherford, J. Kaur, and J. Mestres, Org. Biomol. Chem., 2003, 1, 4235-4241.

13. Solution 1H, 15N NMR Spectroscopic Characterization of Substrate-Bound, Cyanide-Inhibited Human Heme Oxygenase: Water Occupation of the Distal Cavity, Y. Li, R.T. Syvitski, K. Auclair, P. O. de Montellano, and G.N. La Mar, J. Am. Chem. Soc. 2003, 125, 13392-13403.

14. 1H NMR Characterization of the Solution Active Site Structure of Substrate-Bound, Cyanide-Inhibited Heme Oxygenase from Neisseria meningitidis: Comparison to Crystal Structures, Y. Liu, X. Zhang, T. Yoshida, and G. N. La Mar*, Biochemistry 2004, 43, 10112-10126.

15. Solution Structure of Human SUMO-3 C47S and Its Binding Surface for Ubc9, H. Ding, Y. Xu, Q. Chen, H. Dai, Y.Tang, J. Wu, and Y. Shi, Biochemistry 2005, 44, 2790-2799.