Comparison of Response of DRI Oxycodone Semiquantitative Immunoassay With True Oxycodone Values Determined by Liquid Chromatography Combined With Tandem Mass Spectrometry: Sensitivity of the DRI Assay at 100 ng/ml Cut‐Off and Validity of Semiquantitative Value

Abstract

Objective

Oxycodone is a widely used opioid for pain management and patient's compliance with therapy is often monitored by using oxycodone immunoassay. The performance of the DRI oxycodone immunoassay was compared with liquid chromatography combined with tandem mass spectrometry (LC/MS/MS) assay.

Materials and Methods

In 48 urine specimens collected from patients taking oxycodone, urinary oxycodone concentrations were determined using LC/MS/MS and the DRI oxycodone immunoassay for application on the Cobas c 501 analyzer (Roche Diagnostics, Indianapolis, IN).

Results

Out of 48 specimens, 14 specimens showed oxycodone value less than 100 ng/ml, seven specimens had low positive values (between 101 and 165 ng/ml) and all other specimens had values 165 to 1789 ng/ml using the LC/MS/MS assay. The DRI oxycodone assay successfully identified all oxycodone specimens with oxycodone concentrations over the 100 ng/ml. In addition, the DRI assay also showed positive response in 11 out of 14 specimens with oxycodone values less than 100 ng/ml. However, semiquantitative values obtained by the DRI assay did not match with true oxycodone and metabolite oxymorphone concentrations combined obtained by using LC/MS/MS.

Conclusions

DRI oxycodone immunoassay at 100 ng/ml is a reliable immunoassay for analysis of oxycodone in urine.

INTRODUCTION

Opioid therapy is commonly used in treating postsurgical acute pain, other causes of acute pain as well as chronic pain including pain related to cancer. Oxycodone is a semisynthetic opioid narcotic analgesic derived from opioid alkaloid thebaine 1. Oxycodone use in the United States became very popular after release of controlled release preparation OxyContin® and currently it is one of the most commonly prescribed narcotic analgesics in the United States 2. However, like other prescription medications, oxycodone is also abused and such abuse may lead to drug dependence. From 1999 to 2008, sales of opioid drugs including oxycodone have increased by four fold with significant increases in opioid related drug overdose and even mortality 3. Therefore, it is important to have appropriate diagnostic tools to manage the therapeutic use of oxycodone as well as to identify potential diversion of therapy and abuse.

Urine toxicology screening testing is an important standard of care in pain treatment setting offering a reproducible and objective way of monitoring compliance of patients with opioid therapy. Immunoassays are widely used for urine drug screening but opioid drugs such as oxycodone may produce a negative test result using opiate immunoassay due to poor crossreactivity of oxycodone with the antibody used in opiate immunoassays 4. Based on such false negative test result a patient may be inappropriately denied any further prescription for oxycodone. A 40 year old man initially diagnosed with migraine over several years developed chronic daily headache. After several years of using short acting opioid analgesics, the clinician determined that oxycodone 20 mg dosage twice a day was needed for his headache management. After one month the patient complained that the medication was not effective and over next 3 months he was advised to take the medication every 6 hr. Although his pain was under control he routinely called the clinic stating that he finished his medication faster and needed a refill. Potential abuse of oxycodone was suspected and a urine drug test was ordered which was negative for opiates. The clinician dismissed the patient from the clinic based on a negative toxicology report. On his behalf, a family member contacted a toxicologist who informed that oxycodone may cause false negative result in urine opiate drug screen. An aliquot of the original urine specimen was retested using gas chromatography combined with mass spectrometry (GC/MS) in a reference laboratory and the presence of oxycodone at 1,124 ng/ml was confirmed indicating that the patient was compliant with the oxycodone prescription. The original negative result using opiate immunoassay was determined to be due to poor crossreactivity of oxycodone with the opiate immunoassay 5. Therefore, proper testing and selection of the appropriate assay is essential in monitoring therapy in pain management patients.

Oxycodone in urine can be measured by immunoassays or using GC/MS. In recent years, liquid chromatography combined with tandem mass spectrometry (LC/MS/MS) is gaining acceptance as the reference method for analysis of drugs in biological matrix including urine toxicology testing. However, oxycodone immunoassays are widely used in monitoring pain management patients and are often the first assay used in urine toxicology. If a test is positive it may be confirmed by a chromatographic technique but no further testing may be offered if the initial screening test is negative. Abadie et al. commented that the DRI oxycodone immunoassay could be used as a standard technique in detecting oxycodone and its metabolite as none of the 17,069 oxycodone immunoassays results were followed with request for confirmation using gas chromatography‐mass spectrometry 6. Backer et al. commented that the DRI oxycodone assay at a 100 ng/ml cutoff is a highly reliable method for the detection of oxycodone and oxymorphone in urine specimen 7. Nevertheless false negative tests may occur with oxycodone immunoassay when results are compared with a chromatographic method 8. In our laboratory relatively low oxycodone concentrations (between 40 and 150 ng/ml) in urine are routinely observed in some pain management patients using LC/MS/MS. The objective of this study is to determine whether the DRI oxycodone immunoassay could identify such specimens with near cut‐off oxycodone concentration around 100 ng/ml. Moreover, the DRI oxycodone assay is a semiquantitative assay. Therefore, whether such semiquantitative values obtained by the DRI assay could correlate with true oxycodone concentration obtained by the LC/MS/MS method was also investigated.

MATERIALS AND METHODS

For this study, left over urine specimens were used after performing the oxycodone test and reporting results to ordering clinicians. These samples are routinely submitted to our laboratory for monitoring compliance of patients with opioid therapy. Urine specimens were de‐identified prior to further testing and these specimens would have otherwise been discarded.

For analysis of oxycodone one aliquot of urine was analyzed by LC/MS/MS method and another aliquot was analyzed by the DRI oxycodone semiquantitative assay using Cobas c501 analyzer (Roche Diagnostics, Indianapolis, IN) following parameters and instructions of the manufacturer. The cut‐off concentration of this immunoassay was 100 ng/ml. However, regardless of positive or negative test result, the instrument also provided a semiquantitative value. The calibration range of this assay was 100 to 1,000 ng/ml and semiquantitative values over 1,000 were diluted for reanalysis using certified drug free urine.

For LC/MS/MS assay for oxycodone, an aliquot of urine specimen was incubated with beta‐glucuronidase (Sigma Chemical Company, St. Louis, MO) typically for 2 hr at 47°C in order to hydrolyze glucuronide metabolite(s). Then specimens were injected for analysis using an Agilent 1200 series liquid chromatography system including two binary gradient pumps with channel degassers, a thermostatically controlled column compartment, and a refrigerated autosampler (Agilent Technologies, Palo Alto, CA). The analytical column used for reverse‐phased chromatography was an Agilent Poroshell 120 SB‐C18 Column (2.1 × 100 mm 2.7 micron. P.N. 685775–902A). An Agilent 6400 series triple quadrupole mass spectrometer was operated in positive ion mode with dynamic multiple reaction monitoring (MRM) for ion detection and quantitation. Mobile phase A (MPA) was comprised of liquid chromatography grade water with 5 mM ammonium formate plus 0.05% formic acid, and mobile phase B (MPB) consisted of liquid chromatography grade methanol with 0.05% formic acid. The gradient conditions were as follows: starting conditions 10% MPB, at 0.5 min the gradient increases to 15% MPB, with a ramp to 50% at 3 min for analyte elution, followed by a final ramp to 95% MPB at 4 min. The gradient was completed at 5.5 min with a 1.1 min postrun re‐equilibration prior to column switching and next injection. Precursor and product ions along with corresponding deuterated internal standards are listed in Table 1. The deuterated internal standards were purchased from Cerilliant Corporation (Round Rock, Texas). Statistical analysis was performed using linear regression analysis.

Table 1.

Ions Used for Analysis of Various Opioids Along With Ions for Internal Standard

Analyte	Precursor Ion (m/z)	Product Ion (m/z)	Product Ion (m/z)	Internal Standard
Codeine	300.2	165.1	183.1	Codeine‐D3
Dihydrocodeine	302.2	115.1	128	Codeine‐D3
Hydrocodone	300.2	171.1	199.1	Hydrocodone‐D3
Hydromorphone	286.2	128.1	185	Hydromorphone‐D3
Morphine	286.2	157.1	165.1	Morphine‐D3
Norhydrocodone	286.2	128.1	199.1	Hydrocodone‐D3
Normorphine	272.1	152.1	165	Morphine‐D3
Noroxycodone	302.1	227.1	284.1	Oxycodone‐D3
Noroxymorphone	288.1	213	270.1	Oxymorphone‐D3
Oxycodone	316.2	241.1	256.1	Oxycodone‐D3
Oxymorphone	302.1	227.1	284.2	Oxymorphone‐D3
Codeine‐D3	303.2	152.1
Hydrocodone‐D3	303.2	128.1
Hydromorphone‐D3	289.2	185
Morphine‐D3	289.1	152.1
Oxycodone‐D3	319.2	301.2
Oxymorphone‐D3	305.2	230.1

RESULTS

In this study, 48 urine specimens with a wide range of oxycodone concentrations were analyzed using LC/MS/MS and the DRI oxycodone immunoassay. The range of oxycodone concentrations in these specimens as determined by LC/MS/MS varied from 41 to 1789 ng/ml. Out of 48 specimens analyzed, 14 specimens had oxycodone less than 100 ng/ml. Interestingly, 11 out of these 14 specimens showed positive response by the DRI oxycodone assay despite oxycodone concentration below 100 ng/ml (Table 2). In addition, another seven specimens with low positive oxycodone values (between 100 and 165 ng/ml) also showed positive response using the DRI assay (Table 3). As expected all 27 other samples with oxycodone concentrations between 165 and 1789 ng/ml also showed positive responses using the DRI oxycodone assay. Therefore, we did not observe any false negative test result using the DRI oxycodone assay. However, semiquantitative values obtained by using the DRI oxycodone assay were significantly higher in all cases than the true oxycodone concentrations.

Table 2.

Response of DRI Oxycodone Immunoassay to 14 Specimens Containing Oxycodone Below 100 ng/ml as Determined by LC/MS/MS

Urine Specimen ID	Oxycodone by LC/MS/MS	DRI Immunoassay (Semiquantitative Value)
7	67 ng/ml	Positive (322 ng/ml)
17	51 ng/ml	Positive (1841 ng/ml)
19	45 ng/ml	Negative
20	61 ng/ml	Positive (269 ng/ml)
24	74 ng/ml	Positive (789 ng/ml)
29	49 ng/ml	Negative
32	80 ng/ml	Positive (326 ng/ml)
34	72 ng/ml	Positive (175 ng/ml)
36	59 ng/ml	Positive (329 ng/ml)
41	73 ng/ml	Negative
42	56 ng/ml	Positive (577 ng/ml)
43	43 ng/ml	Positive (267 ng/ml)
44	41 ng/ml	Positive (193 ng/ml)
48	51 ng/ml	Positive (106 ng/ml)

Table 3.

Response of DRI Oxycodone Immunoassay to Seven Specimens Containing Between 100–165 ng/ml as Determined by LC/MS/MS

Urine Specimen ID	Oxycodone by LC/MS/MS	DRI Immunoassay (Semiquantitative Value)
2	124 ng/ml	Positive (941 ng/ml)
10	154 ng/ml	Positive (353 ng/ml)
11	149 ng/ml	Positive (909 ng/ml)
14	101 ng/ml	Positive (118 ng/ml)
18	141 ng/ml	Positive (318 ng/ml)
25	138 ng/ml	Positive (497 ng/ml)
28	165 ng/ml	Positive (664 ng/ml)

The 11 specimens that contained oxycodone less than 100 ng/ml but still tested positive by the DRI oxycodone assay were investigated further. These specimens were analyzed to obtain concentrations of oxycodone metabolites including oxymorphone, noroxycodone, and noroxymorphone. All specimens except two (specimen # 34 and 48) showed sufficient concentrations of oxycodone metabolites including oxymorphone which had 103% crossreactivity with the assay explaining why these specimens tested positive despite oxycodone concentration below 100 ng/ml (Table 4). However, specimen # 34 had oxycodone concentration of 72 ng/ml but no detectable level of oxymorphone although the specimen had 1,373 ng/ml of noroxycodone. Similarly, specimen #48 had 51 ng/ml of oxycodone but no detectable level of oxymorphone but the specimen had 603 ng/ml noroxycodone. However, a significant amount of noroxycodone does not explain the positive response by the DRI oxycodone assay due to lack of crossreactivity of noroxycodone with the immunoassay.

Table 4.

Concentration of Oxycodone and Metabolites in Eleven Specimens Which Contained Less Than 100 ng/ml Oxycodone but Tested Positive by DRI Assay

ID	Oxycodone	Oxymorphone	Noroxycodone	Noroxymorphone	DRI Immunoassaya
7	67 ng/ml	147 ng/ml	738 ng/ml	299 ng/ml	Positive (322 ng/ml)
17	51 ng/ml	1089 ng/ml	1748 ng/ml	441 ng/ml	Positive(1841 ng/ml)
20	61 ng/ml	99 ng/ml	ND	ND	Positive (269 ng/ml)
24	74 ng/ml	402 ng/ml	563 ng/ml	101 ng/ml	Positive (789 ng/ml)
32	80 ng/ml	169 ng/ml	956 ng/ml	127 ng/ml	Positive (326 ng/ml)
34	72 ng/ml	ND	1373 ng/ml	ND	Positive (175 ng/ml)
36	59 ng/ml	139 ng/ml	261 ng/ml	ND	Positive (329 ng/ml)
42	56 ng/ml	309 ng/ml	430 ng/ml	175 ng/ml	Positive (577 ng/ml)
43	43 ng/ml	54 ng/ml	600 ng/ml	117 ng/ml	Positive (267 ng/ml)
44	41 ng/ml	99 ng/ml	273 ng/ml	ND	Positive (193 ng/ml)
48	51 ng/ml	ND	603 ng/ml	ND	Positive (106 ng/ml)

^aSemiquantitative value.

ND: None detected.

In addition, three specimens with oxycodone concentration below 100 ng/ml which tested negative by the DRI oxycodone assay were further analyzed for concentrations of oxymorphone, noroxymorphone, and noroxycodone using LC/MS/MS. Specimen #19, #29, and #41 had oxymorphone metabolite concentrations below the detection limit of the assay explaining why these specimens showed a negative response with the DRI oxycodone assay. However, specimen #19 had high amount of codeine (6,400 ng/ml) while specimen #29 had hydrocodone (2,814 ng/ml). Because, codeine or hydrocodone did not crossreact with the DRI oxycodone assay, these two specimens tested negative by the DRI assay where oxycodone concentrations were substantially below the cut‐off level of the assay.

In order to investigate if semiquantitative values matched with combined oxycodone and oxymorphone concentration as determined by the LC/MS/MS method, 24 specimens where semiquantitative values produced by the DRI oxycodone assay were below 1,000 ng/ml were selected because 1,000 ng/ml was the upper limit of linearity of the DRI oxycodone assay. A poor correlation between total concentrations (oxycodone and oxymorphone determined by LC/MS/MS) and semiquantitative concentrations produced by the DRI oxycodone immunoassay was observed (Table 5).

Table 5.

Concentration of Sum of Oxycodone and Oxymorphone Concentrations Compared With Semiquantitative Response Obtained by DRI Oxycodone Assay in 24 specimens With oxycodone Concentrations Between 41 to 165 ng/ml Which also Tested Positive by DRI Assay

ID	Oxycodonea	Oxymorphone	Total	DRI Immunoassayb (bias)
1	238 ng/ml	ND	238 ng/ml	765 ng/ml (+ 221%)
2	124 ng/ml	299 ng/ml	423 ng/ml	941 ng/ml (+ 122%)
7	67 ng/ml	147 ng/ml	214 ng/ml	322 ng/ml(+50.1%)
9.	228 ng/ml	166 ng/ml	394 ng/ml	609 ng/ml (+ 54.5%)
10	154 ng/ml	116 ng/ml	270 ng/ml	353 ng/ml (+30.7%)
11	149 ng/ml	403 ng/ml	552 ng/ml	909 ng/ml (+ 64.6%)
12.	358 ng/ml	ND	358 ng/ml	487 ng/ml (+36.0%)
14	101 ng/ml	ND	101 ng/ml	118 ng/ml (+ 16.8%)
18	141 ng/ml	105 ng/ml	247 ng/ml	318 ng/ml (+28.7%)
20	61 ng/ml	99 ng/ml	160 ng/ml	269 ng/ml (+ 88.1%)
22	344 ng/ml	155 ng/ml	499 ng/ml	981 ng/ml (+ 96.5%)
24	74 ng/ml	402 ng/ml	476 ng/ml	789 ng/ml (+ 65.7%)
25	138 ng/ml	ND	138 ng/ml	497 ng/ml (+260%)
28	165 ng/ml	276 ng/ml	441 ng/ml	664 ng/ml (+50.5%)
32	80 ng/ml	169 ng/ml	249 ng/ml	326 ng/ml (+30.9%)
33	246 ng/ml	91 n/ml	337 ng/ml	550 ng/ml (+63.2%)
34	72 ng/ml	ND	74 ng/ml	175 ng/ml (+136%)
36	59 ng/ml	139 ng/ml	198 ng/ml	329 ng/ml (+ 66.1%)
37	209 ng/ml	114 ng/ml	323 ng/ml	623 ng/ml (+92.8%)
38	213 ng/ml	146 ng/ml	359 ng/ml	603 ng/ml (+67.9%)
42	56 ng/ml	309 ng/ml	365 ng/ml	577 ng/ml (+58.1%)
43	43 ng/ml	54 ng/ml	97 ng/ml	267 ng/ml (+175%)
44	41 ng/ml	99 ng/ml	140 ng/ml	193 ng/ml (+37.8%)
48	51 ng/ml	ND	51 ng/ml	106 ng/ml(+108%)

^aOxycodone and its metabolites were determined by using LC/MS/MS.

^bSemiquantitative value obtained by the DRI oxycodone assay.

DISCUSSION

Pain management usually includes urine drug testing as a measure of compliance with prescription drugs. Unlike workplace drug testing where guidelines are well defined by SAMHSA, guidelines for drug testing in pain management are not as well defined. However, the American Society of Interventional Pain Physicians is developing guidelines and a few states have proposed or adopted guidelines for opioid therapy in patients suffering from chronic pain 9, 10, 11, 12. Gas chromatography/mass spectrometry or LC/MS/MS based assays have high sensitivity and specificity and should be used for urine toxicology analysis. Nevertheless immunoassays are routinely used for urine toxicology analysis including monitoring compliance of patients with narcotic analgesic during pain management. Ramirez‐Fernandez Mdel et al. recently described a method for the determination of 23 opioids (including oxycodone), cocaine and metabolites in urine using ultraperformance LC/MS/MS. The lower limit of quantitation for oxycodone in the assay was reported to be 3 ng/ml 13. However, widely used DRI oxycodone immunoassays are available at 100 ng/ml or 300 ng/ml. Therefore, if more sensitive DRI oxycodone assay at 100 ng/ml cut‐off is used in monitoring compliance of patients taking oxycodone, it is important to ensure that an immunoassay should be able to detect a low positive value near the cut‐off value. Our results indicated that the DRI oxycodone assay showed positive response with all specimens with oxycodone concentration of 100 ng/ml or more as determined by the LC/MS/MS assay. In addition, 11 out of 14 specimens with oxycodone concentration less than 100 ng/ml also tested positive with the DRI oxycodone assay again indicating excellent sensitivity of this assay to detect even low levels of oxycodone if present in the specimen.

The package insert of the DRI oxycodone assay claimed that this assay is very specific for oxycodone and its metabolite oxymorphone but exhibits very low crossreactivity with other opioids. In addition, this assay also has low crossreactivities with other oxycodone metabolites such as noroxymorphone (<0.1%) and noroxycodone (<0.1%). Because the DRI oxycodone assay can identify the oxycodone metabolite oxymorphone (103% crossreactivity), positive response is expected even if oxycodone concentration is below 100 ng/ml if the combined oxycodone and oxymorphone concentration exceed 100 ng/ml. This phenomenon explains why some specimens with oxycodone concentration substantially below 100 ng/ml were tested positive using the DRI oxycodone assay.

Backer et al. reported that the DRI oxycodone assay was very specific for oxycodone where no crossreactivity was observed even with codeine concentration over 75,000 ng/ml and hydrocodone concentration over 7,500 ng/ml 7. However, both false positive and false negative results using the DRI oxycodone assay have been reported 8. A presumed hook effect in semiquantitative DRI oxycodone assay has also been reported. The authors also questioned the semiquantitative utility of the assay 14.

Prozone or hook effect is observed when very high amount of an analyte is present in the sample but the observed value is falsely lowered. This type of interference is observed more commonly in sandwich assays. The mechanism of this significant negative interference is the capability of high level of an analyte (antigen) to reduce the concentrations of ‘‘sandwich’’ (antibody 1: antigen: antibody 2) complexes responsible for generating the signal by forming mostly single antibody: antigen complexes. Hook effect has been reported with assays of a variety of analytes such as β−hCG, prolactin, calcitonin, aldosterone, cancer markers (CA 125, PSA) etc. but less commonly observed in a competition immunoassay. The Best way to eliminate hook effect is by serial dilution.

The DRI oxycodone assay is specific for oxycodone and oxymorphone. However, when we evaluated 24 specimens which tested positive for the DRI oxycodone assay for semiquantitative values produced by the assay with true combined oxycodone and oxymorphone values determined by the LC/MS/MS method, we observed a poor correlation. In general the DRI oxycodone assay showed significant positive bias although negative bias was also observed in few specimens. However, a trend where higher values obtained by the DRI immunoassay in general indicated a higher amount of oxycodone plus oxymorphone concentration in urine was also observed. In conclusion, the DRI oxycodone immunoassay at 100 ng/ml cut‐off is a reliable assay to detect oxycodone concentration in urine with no false negative results. However, in order to determine true concentrations of oxycodone in urine use of a specific chromatographic method such as gas chromatography combined with mass spectrometry or LC/MS/MS is recommended.