BACKGROUND OF THE INVENTION

Field of the Invention

The present invention relates to methods and composi- tions for testing pancreatic exocrine function. More spe¬ cifically, it relates to the use of the vitamin B ₁₂ ingestion system as a means of measuring pancreatic exocrine function. Prior Art

The human pancreas contains endocrine portions and an exocrine portion. The endocrine portions actually consist of a multitude of small islets that synthesize and release hormones into the blood. The exocrine portion carries the endocrine islets and makes up the bulk of the pancreas. When functioning properly, the exocrine portion synthesizes a number of digestive enzymes such as amylase, trypsin, chymotrypsin, elastase, lipase, and others. These enzymes are released into the upper part of the small intestine and are essential for proper digestion and nourishment.

On occasion the exocrine portion of the pancreas may be caused to malfunction. Such malfunction may be due to damage caused in a variety of ways, including chronic al¬ coholism, trauma, and pancreatic cancer. Due to the fact that the pancreas is located deep in the abdomen it is extremely difficult to evaluate clinically. X-rays and scans of the pancreas are discernably abnormal in only a relatively small proportion of patients with actual pan¬ creatic exocrine insufficiency.

Several tests of pancreatic exocrine function are currently available. However, such tests are time con- suming, cumbersome and inconvenient. One prior art test involves measuring the amount of fat in feces. This fecal fat test is based on the fact that fecal fat is increased in the absence of the pancreatic exocrine enzyme lipase which is required for the digestion and absorption of fat in the intestine. To accomplish this test the patient

must save all of his feces for 72 hours and the faces must then be analyzed for fat in the laboratory. This test is unpleasant for the patient and for laboratory personnel and, even then, it is not available at many hospitals. Furthermore, this fecal fat test lacks specificity to pan¬ creatic exocrine insufficiency, since it also provides ab¬ normal results in a variety of intestinal diseases such as Crohn's disease and sprue.

Another prior art test, duodenal intubation, involves placing a tube into the upper small intestine via the mouth and stomach and then collecting pancreatic juice which is subsequently analyzed for volume, bicarbonate and pancreati enzymes. This test is time consuming, expensive, incon¬ venient and unpleasant for the patient, and again, it is available at only a relatively small number of medical centers. It is therefore seen that there is presently no convenient easily applied test for pancreatic exocrine function.

It has been noted, for example, employing standard Schilling tests that B- _j ^ is malabsorbed by about 50% of patients diagnosed to have pancreatic exocrine insufficien¬ cy. However, B^2 malabsorbtion by itself is not a reliable indication of pancreatic exocrine dysfunction as B^ ab¬ sorption is not abnormal in more than about one-half of patients with such disfunction, and further ¹ , as B-, ₂ ^aD_ sorption is also abnormal in cases of pernicious anemia and also as a result of a number of intestinal diseases. The Schilling test, noted above, utilizes labelled ^B 12' f° ^r example, B-^ including a radioactive cobalt isotop to measure and evaluate absorption of B-^ b the digestive system. A modified or dual labelling Schilling test is commercially available, which uses two differently labelled forms of B, ₂ r ^on e °f which is bound to intrinsic factor, and the.other of which is unbound. However, such tests utilizing labelled have at no time been used to inde¬ pendently evaluate pancreatic exocrine function.

For many years it was believed that ingested vitamin B^2 was bound in the form of a complex to a protein known as intrinsic factor (IF) present in the gastric juices of the stomach. It was further believed that this IF-B-^ com- ^' plex remained intact until it attached to specific receptors located at the end of the small intestine at which receptors the ^waΞ unbound and absorbed into the body. It has now been determined that this model for ingestion is in¬ correct and that use of the correct B- _j ^ ingestion model can now serve as the basis of accurately evaluating pancreatic exocrine function. BRIEF DESCRIPTION OF THE INVENTION

As used throughout this application, the term "B- _j ^" shall mean and include "vitamin B-^" and the various "cobalamin" compounds and "corrinoids" which are forms of or antecedents of vitamin B-^.

The present invention is based on a recently discovered model of B-, ₂ absorption into the system. As already noted, it had been previously believed that ingested B, „ formed a complex with and was bound to the protein designated as "intrinsic factor" (IF) which is normally present in the gastric juices. It has now been determined that the mechan¬ ism for absorption is a two-step process. In this two- step process B-^ ₂ initially binds preferentially with an- other protein, designated as "R protein," which is also normally present in the gastric juices. This complexing of B-^2 with R protein is overwhelming, even in the presence of IF protein. On the order of about 98 to 99% of the B ₁₂ reacts with normally available R protein to form a complex, which is herein designated as "R-B-,2 _' " ^In situations in which B]_2 is already complexed with IF, available R protein will remove the B-, ₂ from the IF to overwhelmingly form ^R ~ ^B 12_ After R-B^ ₂ complex enters the small intestine it is then degraded by enzymes from the exocrine pancreas, such as trypsin, chymotrypsin and elastase. Following this degradation the B- _j _ ₂ can then bind with IF to form an

IF-B ₁₂ complex from which B, ~ is eventually ingested via receptors at the end of the small intestine.

It has also been recently discovered that there are vitamin B-, ₂ analogues, which respond to many tests in a manner similar to B, ₂ , but which do not support the same bodily functions as B-, ₂ _ The relationship between B- ₂ and analogue B, ₂ , as well as the accurate testing of B^ ₂ in the presence of B. „ analogue is set forth in great detail in pending U. S. Patent Application Ser. No. 893,524 also by Robert H. Allen and assigned to the same assignee as the present application. It has also been discovered that while R protein will complex with either B- _[ _2 or ana¬ logue B^ ₂ , that IF is specific to and will only bind with true B-,.-.. Discovery of the two-step B-^ ingestion process, of B, ₂ analogue, and of the ability of B, ₂ analogue to bind with R protein, but not with IF, serve as the basis for the present invention.

It is postulated that since B. „ initially preferen- tially binds with R protein the digestive system to form R-B-,2 complex, and that since the ^R ""B _] _2 complex is stable, until acted upon by enzymes produced by the exocrine por¬ tion of the pancreas, that a method of accurately deter¬ mining the ingestion of initially B-^ bound to R reflects on the availability of digestive enzymes from the exocrine portion of the pancreas. It is further postulated that the availability of digestive enzymes from the exocrine portion of the pancreas reflects on the functioning of the exocrine portion of the pancreas. This theory is made the basis of the test ^" of the present invention, which, by providing two differently labelled B-^ ₂ compounds and manipulating the digestive en¬ vironment of the patient undergoing evaluation provides a test for pancreatic exocrine function. While the present invention may simplistically be considered a "dual labelled Schilling type test, it requires steps and materials not

- URE

OMPI

^Z°v

normally used in a Schilling test .

For purposes of this specif ication a f irst labelled k .

B _j _ ₂ compound designated as "B- _j? " ^an< 3 ^a second, but differ-

^■ x k ently labelled B, „ compound, designated as "B^" are re- k k k . quired. ^~ - 2 ^an d ^B i2 ^ma y ke, ^ ^or example, radioactive

( ⁵⁷ Co) B ₁₂ and radioactive ( ⁵⁸ Co) B, ₂ . ^otner isotopes or means of labelling B-. - , may also be used.

In the practice of the present invention, as a test for exocrine pancreatic function, a patient ingests, either as a single mixed composition or as a series of several

-k k compositions: (1) vitamin B- _j _2 bound to intrinsic factor, k k k and hereinafter designated as "IF-B- _j ^;" (2) vitamin B-^ bound to R protein, and hereinafter referred to as "R-B^;" and (3) a substantial amount of B-,2 analogue. In preferred embodiments a fourth component containing unbound intrinsic factor, hereinafter "IF," is also included.

Subsequently, as detailed below, the amount and ratio k k k of B_2 and B12 absorbed, as assayed by measuring the amounts in the urine using the Schilling procedure is in- dicative of the function of the exocrine pancreas.

In the general practice of the process of the present invention the patient undergoing test is given an injection of B-,2' or is otherwise treated to remove free binding pro- k k k tein from his blood, and caused to ingest IF-B- _j _2' ^R-B 12 and B, ₂ analogue. Subsequently his urine is collected and k k k the amount and ratio of B^ ₂ ^an d B^2 measured. Normal exo¬ crine pancreas function will be attributed to patients who excrete B-^ nd B-- in substantially the same proportion as it was ingested. Abnormal pancreatic exocrine function will be attributed to patients who excrete substantially k k -k more ^~ -γ- than B^ i relation to the proportions ingested.

The following mechanism is postulated for this test. The vitamin B-^ injection.which the patient is given binds the subject's then available binding proteins in the blood and serum so that that binding material is not available k k k to bind with the ingested B- _j _2 and B^ samples, both of

which can thus be excreted in the urine in substantial a- mounts. ' In the digestive system the ingested B ₁₂ analogue binds with substantially all of the R protein naturally in the stomach, thus making the natural R protein substantiall

** unavailable for competition with the IF ^-B i2 so that the IF k k is not replaced with R in the stomach to form R-B^.

In evaluating the results of the test of the present invention, subjects with normal pancreatic exocrine func¬ tion are found to absorb and excrete significant amounts k k k - of both B12 ^an B12 in about the same proportion as the amounts of B _j ^ ^{and B} ]_2 originally ingested. This is due to the fact that the IF-B 2 ^is readily available for ab¬ sorption without interference from R protein (which has been bound by the ingested B- _j _2 analogue) , while the pan- creatic exocrine enzymes in the normal system cause the

R-B *12 to be degraded so that the Bj*.2 can then bind with

IF in the system for absorption via the receptors. How- k k it _• ever, not much of either the B^2 ^{or B} 12 ^rerna ins in the body, since the normal binding protein has been bound by a vitamin B-^ injection.

Utilizing the test of the present invention, patients with pancreatic exocrine insufficiency will absorb and ex- crete substantial amounts of B- _j _ ₂ which was initially bound with IF. However, using this test, they will absorb and excrete very little B^2 _* This is due to the fact that, as ingested the B- _j ^ was initially bound to R protein. There¬ fore, the absence or insufficiency of pancreatic exocrine k -k enzymes to digest the R-B^2 does not allow the B^ to be released from R protein for binding with IF and for absorp- tion as IF-B _j _2- Thus, during the practice of the present it invention, low excretion of B-^ signals a lack of pancreati exocrine enzyme in the system, and thus a malfunctioning exocrine pancreas.

In the absence of pancreatic disease, patients having intestinal diseases, which interfere with B _j 2 absorption, absorb and excrete decreased, but essentially proportional

amounts of B. „ and B, _, to the amounts ingested.

Patients with both pancreatic exocrine disease and intestinal disease will also absorb and excrete decreased amounts of both but the amount of decrease for

5. the B^2 i disproportionately greater than the amount of

** decrease for the B- ₂ .

In preferred embodiments, the patient undergoing test is also caused to ingest unbound IF. This assures the

* availability of IF to bind with B^2 when and if it becomes 0 available from degraded R-B 2« This assures the test a- gainst erroneous results should the patient fail to have sufficient IF in his gastric juices.

Other embodiments of the present invention include mix- k k k tures of compositions containing IF-B-j_2 ^R_B i2 ^an< 3 ^B i2 ^ana ~ 5 logue, with and without unbound IF. Methods of making k k k

IF-B-J ₂ ^an d the R-B ₁₂ complex constitute yet other embodi¬ ments of the present invention.

The foregoing and other objects, features and advan¬ tages of the invention will be apparent from the following more particular description of preferred embodiments of the invention. DETAILED DESCRIPTION OF THE INVENTION

The following is set forth as a prelude to describing the process and compositions of the present invention. They provide details of certain experiments' which were carried on and which led to the hypothesis upon which the present invention is based. Additionally, they provide evaluation of that hypothesis and further indicate the degree of specificity and sensitivity of the tests of the present invention.

As already noted, the present invention is based upon a BJ 2 absorption model which proposes that ingested B-, ₂ i first bound to R in the gastric juices. It is further pro¬ posed that only after pancreatic proteases act is the R ^-B i2 . complex degraded so that B-, ₂ can bind with IF for absorption into the system. Based on this model normal subjects should

absorb B-^2 originating as intrinsic factor-B ₁₂ originating as R protein-B ₁₂ i ⁿ equivalent amounts. Furthermore, rati of excreted B-, ₂ originating as intrinsic factor-B-^ ^{to B} i2 originating as R protein-B- _j _2 should be close to the ratio ^' at which they were ingested. Subjects with pancreatic exo crine insufficiency should absorb B-^ originating as intri sic factor-B ₁₂ normally, but should malabsorb B ₁₂ originat as R protein-B- _j _2- Thus, if given substantially equal a- mounts of IF-B ₁₂ ^a d R-B- _j ^ they should have high ratios for B^2 originating as intrinsic factor-B-j_2 ^{to R} protein—B^- Subjects with intestinal disease would often malabsorb B- _j ^ originating as both intrinsic factor-B^ ^an R protein-B-j^ but such malabsorption should be equivalent. ig-

ate all designate the latter complex.

Materials and Methods TB-m ₁ - ₂ - .(l12 -.5 uCi/nmmoll) and { ,5 ^D 8 ^O ,Co) B (4.0 uCi/nmol) were obtained from Amersham-Searle Corp. , Arlington Heights IL. The ( ⁵⁸ Co)B- _] _2 was obtained in the form of capsules such as those utilized in the Dicopac Schilling Test Kit. Each capsule was opened and the contents dissolved in 5.0 ml of 0.15 M NaCl solution and centrifuged at 10,000 x gravity for 15 minutes at a temperature of 4°C. Then the supernatant liquid containing the ( ⁵⁸ Co)B- _| 2 ^was stored at -20°C. Crystalline B ₂ was obtained from Sigma Chemical Company, St. Louis, MO., as CN-cobala in assayed spectro- photometrically, and utilized to dilute the ("coJB-^ ^to a specific activity of 2.5 uCi/nmol. analogue in the form of cobinamide was prepared and isolated as described by Allen et al J. Clin. Invest. 60:1381-1392, 1977, except that cobinamide and 13-epicobin- amide were not separated from each other by paper chro a- tography.

obtained from Sigma Chemical Company.

Crystalline hog elastase was obtained from P-L Bio- chemicals, Inc., Milwaukee, I. Pancreatic extract was obtained from Kremers-Urban

Company, Milwaukee, WI. , in the form of capsules designated as (KU-ZYME HP) .

Homogenous human intrinsic factor, hog R protein, rabbit anti-human intrinsic factor serum, and human anti- intrinsic factor blocking antibodies were obtained as des¬ cribed previously by Katz et al, J. Clin. Invest. 53:1274- 1283, 1974.

Human saliva served as the source of human R protein. Human duodenal juice was collected with a dual lumen duodenal tube. The distal opening was positioned at the ligament of Trietz under fluoroscopic control with the second opening being located 12 cm proximally. Pancreatic secretion was stimulated by perfusing a mixture of essen¬ tial amino acids and glucose through the proximal opening at a rate of 6 l/min and samples were collected through the distal opening. The perfusate used to stimulate pan¬ creatic secretion contained the following, in g/liter: lysine, 5.5; tryptophan, 2.0; phenylalanine, 5.5; methi- onine, 8.5; threonine, 3.8; leucine, 7.9; isoleucine, 3.7; valine, 6.1; and glucose, 50.

Vitamin B-^ ₂ binding assays were performed as described by Allen et al, J. Biol. Che . 248: 3660-3669 (1973).

The transfer of ( f ^r om R protein to intrinsic factor was performed at pH 8.0 and was based on the pre¬ cipitation of ( ⁵⁷ Co)B ₁₂ intrinsic factor with rabbit-anti- intrinsic factor antisera in 40% (NH ₄ ) ₂ S0 ₄ as described by Allen et al, J. Clin. Invest. 61: 47-54, 1978.

Trypsin was assayed, as described by Erlanger et al, Arch, Bioche . Biophy. 95: 271-278, 1961, with benzoyl arginine p-nitroanilide, and the results were expressed as ug equivalents of hog trypsin which was used as the standard,

The following studies are indicative of the fact that pancreatic enzymes are required in order to transfer B- _j _ ₂ from B; _j _2 ^-R complexes to B ₁₂ ^-IF complexes.

The results of in vitro studies with human R protein and purified pancreatic proteases are presented in Table 1 and indicate that 50% of the BJ2 ^was transferred from human R protein to human intrinsic factor at concentrations of hog trypsin and bovine chymotrypsin of 80 and 1200 ug/ml, respectively. Hog elastase was the least effective in transferring B^2 from R protein to IF, with only 11% trans¬ fer occurring at a hog elastase enzyme concentration of 2250 ug/ml. However, these proteases appear to act syner- gistically since an equal mixture of all 3 enzymes effected 50% transfer at a total protein concentration of 63 ug/ml (concentration of each enzyme 21 ug/ml) .

The results of identical studies performed simultane¬ ously with hog R protein are also shown in Table 1. At enzyme concentrations of 2250 ug/ml only 11, 4 and 0% of B12 was transferred from hog R protein to human intrinsic factor with hog trypsin, bovine chymotrypsin, and hog elastase, respectively. Only 6% transfer of B _] _2 occurred with an equal mixture of all 3 enzymes at a total enzyme concentration of 2250 ug/ml. These results indicate that hog R protein-B]_2 i ^Ξ 50-100 fold more resistant to the k action of pancreatic proteases than is human R protein-B^- Table 2 shows the results of similar studies performed with a freshly prepared solution of hog pancreatic extract (KU-ZYME HP) and freshly collected samples of human duo¬ denal juice obtained before and after pancreatic secretion was stimulated by the duodenal perfusion of essential amino acids. As shown in Table 2, all three solutions caused 50% transfer of B- _j ^ from human R protein to human intrinsic factor when they were present at -1 to 5% of their full strength. The sample of human duodenal juice obtained afte pancreatic stimulation was approximately 5 times more po¬ tent than the unstimulated sample, while the solution of

( ⁵⁷ Co)B ₁₂ FROM R PROTEIN TO INSTRINSIC FACTOR

Percent Transfer of (57co)B ₁₂ from Human R Protein to Human Intrinsic Factor by Various Pancreatic

Enzymes enzyme added trypsin chymotrypsin elastase all three

(ugml) (*) (*) .%) . %)

0 0 0 0 0

5.5 5 0. 0 5-7

16.5 15 0 0 19

55 39 0 0 6

170 72 -] 0.3 82

550 91 30 5 96

2250 9 70 11 97

Percent Transfer of (57co)B ₁₂ ^{frora Hc} € ^R Protein to Human Intrinsic Factor by Various Pancreatic

Enzymes

enzyme added trypsin chymotrypsin elastase all three

(ugml) {% ) - (50 .% ) (.% )

0 0 0 0 0

5.5 0 0 0 0

16.5 0 0 0 . 0

55 0 • 0 0 0

170 0 0 0 0

550 1.4 0.7 0 1

2250 13 4 0 5-5

Assays were performed in a volume of 0.45 ml of 0.013 I. Tris (Hydroxy- methyl) Aminomethane-HCl pH 8.0. 0.187 NaCl, containing enzyme as indicated, 26.67ug of bovine serum alburrάn, and 1 pmol of (57co)B2 ₂ prebound to 1 pmol of R protein. After incubating for 30 minutes at 37°C, 1.5 pmol of Human Intrinsic Factor was added in 0.05 ml and the reaction was incubated for another 30 minutes at 37°C. The sairples were then cooled in ice water for 5 minutes and 0.1 ml of rabbit Anti Human Intrinsic Factor anti-serum was added. After an additional 30 minutes, 0.5 ml of H ₂ 0 saturated vάth (NH4) ₂ Sθ4 was added, followed by centrifugation at 20,000 x g at 4°C for 30 minutes. A portion, 0.5 ml, of the supernatant was then assayed for 7co)B 2- This reflected the amount of (57co)B ₁₂ still bound to R since any (57Co)B _j2 transferred to Intrinsic Factor was precipitated under the assay condit ons.

Table 2

ABILITY OF CRUDE PANCREATIC ENZYI-ES TO EFFECT THE TRANSFER OF (57co)B. 12 FROM R PROTEIN TO .INTRINSIC FACTOR

Percent Transfer of (5<Co)B ³ 1 ₁ 2 from Human R Protei to Human Intrinsic Factor

solution present* Hog pancreatic extract _Human duodenal juice

{% of full strength) itimulated stimulate

0 0 0 0

0.155 0 2

0.225 0

0.335 0.5 9

0.66 16 32

2.20 61 ■ 2.75 15 90

5-50 52

6.60 89 11.0 83 91 22.0 93 3-5 96 99 50.0 98 66.0 98 99 86.0 99 99 100

Table 2 Continued

Ability of Crude Pancreatic Enzymes to Effect the Transfer of.( ⁵⁷ Co)B 12 From R Protein to Intrinsic Factor

Percent Transfer of (57co)B ₁₂ from Hog R Protein to Human Intrinsic Factor

solution present ^5? Hog pancreatic extract Human duodenal juice

{% of full strength) unstiiπulated st__mulated

0 0

0.225 0

0.66 0

1.30

2.20 0.3

5-5 3.3

6.60 1.7

11.0 0 8.7

22.0 9 0 16

33.0 22 . 3.7 30

51.0 21

66.0 10 , i)0

87.0 13 ^5

90.0 39

Assays were performed as described in Table 1.

# % of full strengths(amount of sample assayed-^initial incubation volume) xl00_. For example, if 0.2 ml of Human Duodenal Juice was added to this assay the ^" % of full strength" would equal Q-ff, ¹ " ¹ , x 10055 ^' = Λ Λ%

U. > mi All volume differences associated with the various amounts of solution present were corrected with 0.15 M NaCl.

hog pancreatic extract was intermediate in this regard. The three solutions had the same relative potencies when tested with hog R protein-B-, ~ (Table 2) , but only about 13-46% of the B ₂ ^was transferred to human intrinsic factor when they were tested at 89% of their full strength. These studies support the concept that hog R protem-B^ ^1S mo e resistant to the action of pancreatic enzymes than is human

R protem-B-j_2'

Table 3 shows the results of an experiment in which multiple serial samples of duodenal juice were collected from a normal subject before and after pancreatic secretion was stimulated by the duodenal perfusion of essential amino acids. Trypsin activity and the ability to transfer B- _j ^ from R protein to intrinsic factor increased in parallel after pancreatic stimulation. This provides additional evidence for the concept that the transfer of B- _j ^ from R protein to. intrinsic factor is a specific function of pan¬ creatic enzymes.

It has been found that dog R protein is similar to hog R protein, when it is used in place of hog R protein in experiments similar to those in Tables 1, 2, and 3. It has also been found that rat IF and hog IF are similar to human IF when used in the practice of the present invention In the preferred process of the present invention, the subject to be tested has a digestive tract substantially free of undigested food or unprocessed waste materials. This is accomplished, for example, by fasting the subject for about 8 to 12 hours. Then the subject is caused to orally ingest the following, either separately or in com- bination:

1) One labelled form of B^2 bound to intrinsic factor (B12-IF) ;

2) A second labelled form of B^2 bound to R protein (B-, ₂ ~ ^R )

3) An unlabelled B, ₂ analogue that binds only to R protein and

4) Free intrinsic factor (IF) (optional) .

OM Λ. IP

CORRELATION OF TRYPSIN ACTIVITY AND THE ABILITY TO TRANSFER (5 co)B 1, 2

FROM R PROTEIN TO IF IN HUMAN DUODENAL JUICE BEFORE AND AFTER

PANCREATIC STH€JLATION

Percent transfer of ( ⁵⁷ Co)B ₁₂ from R ppotein to

Time after pan- Human Intr insic FactorΦ creatic stirπulation Trypsin Human R Hog R

(minutes) ugml ^' %) m

-10 37 7 0

- 5 54 11 0

0 35 1.4 0

+ 5 97 19 0

+10 272 82 10

+15 360 89 14

+20 271 83 9

+25 189 76 . 1.7

+30 129 60 Q

+35 138 54 0

+4θ 145 52 0

+45 145 57 0

+50 123 38 0

+55 172 58 0

+60 . 132 41 0

+65 135 ^• 46 Q

+70 ^• 166 50 0

+75 120 42 0

+80 95 3θ ^' 0

+85 ni 32 0

Assayed as described in specifications

* Assayed as described in Table 1. 0.020 ml and 0.400 ml of each sample were used for the human R and hog R assays, respectively.

These test solutions were prepared by combining 1 ml of H ₂ 0 containing 0.2 n ol of ( ⁵⁷ Co)B^2 ( ² - ⁵ uCi/nmol) with 0.6 nmol of purified human intrinsic factor present in 4.0 ml of 0.01 M potassium phosphate, pH 7.5, containing 0.15 M NaCl and 50 ug/ml of bovine serum albumin. Alter¬ natively, human gastric juice which contains both IF and R protein could be utilized after incubation with suitable B _] _2 analogue, such as cobinamide, B- _j ^ (bde-OH) or (3,5,6- Me3 benzimidazole)B;j_2 in 1 to 1,000,000 fold excess with respect to R protein. These B ₁₂ analogues bind to R, but not IF. Thus when ( ⁵ 7co)B ₁₂ i ^s subsequently added it is bound only by IF since all of the R is already bound. In a separate tube, 5.0 ml of 0.15 M NaCl containing 0.2 nmol of ( °Co)B 2 (4.0 uCi/nmol) was combined with 0.3 nmol of R protein present in 5.0 ml of 0.01 M potassium phosphate, pH 7.5, containing 0.15 M NaCl and 5.0 ug/ml of bovine serum albumin. The IF and R protein solutions were incubated at room temperature for 30 minutes and then cooled to 4°C. Cobinamide, 200 nmol in 1.0 ml of 0.15 M NaCl, was added to the R protein solution and after incubating for 30 min¬ utes at 4°C the intrinsic factor and R protein solutions were combined and stored at -20°C. After thawing, it was found that greater than 95% of the ( ⁵⁷ CO)B£2 an (58CO)B ₂ were still bound by IF and R protein, respectively, based on assays employing charcoal adsorption and _< precipitation with specific antisera.

Test solutions were administered orally to subjects who had been fasted for about 8 to 12 hours. An intramus¬ cular injection of 1 rag of nonradioactive (unlabelled) B-, ₂ was given at substantially the same time as the oral test solution. Subsequently, about two hours after the test was begun, the subject was allowed to end his fast. The subject's excreted urine was then collected for 24 hours after the administration of the test solution and was assayed for ( ⁵ 'Co) and ( ⁵ ^co) with a Beckman Gamma 300 System, Beckman Instruments, Inc., Fullerton, CA.

exocrine function on the ability to absorb B-,~ bound to human R protein or hog R protein, versus that of B- _j ^ bound to human intrinsic factor, was tested by performing a series 5 of dual label Schilling type tests. Each orally adminis¬ tered test sample contained 0.2 nmol of ( ⁵⁸ Co)B _] _2 bound to either human R protein or hog R protein, 0.2 nmol of

^'*" ( ⁵⁷ CO)B ! bound to human IF, 0.4 nmol of human IF and 200 nmol of nonradioactive cobinamide, a B- _j ^ analogue. Cobina-

10 mide analogue binds to R protein but does not bind with IF. When ingested it serves to prevent endogenous R protein, normally present in the gastric juices, from removing the ( ⁵⁷ co)B ₁₂ from human IF within the gastrointestinal tract. Nonradioactive B.„, l.mg, was injected intramuscularly in

15 conjunction with the oral ingestion of the test solutions. This injection served to saturate B-^-binding proteins in the plasma, thus allowing substantial urinary excretion of absorbed ( ⁵⁷ CO)B£ ₂ ^an d ( ⁵⁸ Co)B ₁₂ which could then be meas ^¬ ured in 24 hour urine collections.

20 The results of these tests when performed on a subject with known pancreatic exocrine insufficiency are shown in Table 4 and are consistent with the _in vitro experiments pro-

25 l than Bj_2 originating as human R protein-Bj_2# 3) hog tryp-

* sin corrected the malabsorption of B^2 originating as human R protein-B-j_2 to a much greater extent than it cor- 30 rected the malabsorption of B, ₂ originating as hog R pro- tein-B- _| 2- ^an d 4) hog pancreatic extract was more potent than hog trypsin in correcting the malabsorption of B- _j ^ originating as hog R protein-B-^. Because of its greater availability and resistance to pancreatic enzymes, hog R 35 protein was utilized for the remainder of the preferred embodiments presented below.

Table 4

Comparison of the absorption of ( Co)B* ₂ hound to Human and Hog R protein (R), versus that of ( 7co)B|2 bound to Human intrinsic factor (IF) in dual label pancreatic tests performed on a patient with ^• pancreatic exocrine insu ficiency.

Percentage of acLministered B]_2 excreted in urine

R protein Addition ^"1" W-B ^~ [ ^~ > R-B12 ΣF-3f R-BΪ2 .% ) '

Human None 19.1. 1.5 12.7

Hog None 23-6 0.6 39.3

Human Trypsin* 25.8 19-9 1.3

Hog Trypsin* 16.0 1.8 8.9

Hog Hog 14.3 6.3 2.2 pancreatic extract^

.

⁺ Given orally __π_τediately following the solution containing labelled B ₁₂ .

Crystalline Hog trypsin, 50 mg, freshly dissolved in 50 ml of O.OOIM HCl.

^ Four pancrelipase capsules (KU-ZYME HP) .

Following are details of how the relative absorotion k k k k of B12 originating as human IF-B ₁₂ originating as hog R pro- k tein-B^2' serves as a diagnostic test for pancreatic exocrine insufficiency. The results of dual label Schilling type tests employing human IF-B^2 and hog R protein-B^ performed on normal subjects, and patients with pancreatic and other diseases are shown in Table 5. In 11 normal subjects the percent of orally administered B- _j ^ i 24-hour urine samples ranged from about 15.4% to 26.7% (mean 21.9%) for B^ originating as IF-B12 and from about 10.0% to 23.1% (mean 17.3%) B12 ^{for B} 12 origi ^Tne ratio of urine excreted 1.0 to 1.8 (mean 1.3) . The fact that normal subjects usually ab- k k k k k sorbed slightly more B]_2 originating as IF-B12 than B12 originating as hog R protein-B_L2 i ^s consistent with the above ι_n vitro observation, that human pancreatic juice is somewhat limited in its ability to effect the transfer of B12 from hog R protein to human IF.

In 10 patients with pancreatic disease severe• enough to cause steatorrhea (malabsorption of fat) , as documented independently by tests for elevated fecal fat contents, values for B]_2 originating as IF-B12 were close to values k k of B12 for normal sub ects and ranged from about 9.8% to 23.3% (mean 16.5%). The reason for the slight reductions of seen in some patients is unknown although they could be due to incomplete urine collections, or minor abnormalities of intestinal pH, bac¬ terial overgrowth, or intestinal mucosa integrity. How- k ever, for these same patients the values for B]_2 origina-

_* k ting as hog R protem-B_L2 were uniformly and markedly reduced, ranging from about 0.5% to 2.2% (mean 1.2%). k k k

Values for the ratio of urine excreted B12 ^{to B} 12 were markedly elevated and ranged from about 6.9 to 46.2 (mean 19.4). The values for B12 and the ratio of B12 to BΪ2 did not overlap with the values for any of the normal subjects. The degree of elevation of the ratio of B^2 to B12 appeared

Tables (Continued)

Percentage or orally _-_π_nl8ter_d B ₁₂ in 2k i,__r urine

Subject w Ape Sex Fecal Fat* Dlacaae -F-- ^~ - ^' \2 IF-b_5R--Ϊ ₂

Yr (g/24 llr) (X) (ϊ)

Patients with pancreatic disease without steatorrhea

22 49 M 3 chronic pancreatitis, Etoh, pancreatic calcification, 25% 14.4 5.8 2.5 pancrcatectomy, moderate .lbroβla

23 25 6 acute pancreatitis, chronic steroid therapy for asthma 30.5 20.9 1.5 24 36 2 chronic pancreatitis, Etoh 23.0 19.2 1.2 mean (22.6) (15.3) (1.7) range (14.4-30.5) (5.8-20.9) (1.2-2.5)

Patlenta with other gastrointestinal diseases

25 18 F 39 Crohn's disease, esenterlc infarction, 951 am. bowel reesseecct.iloonn 1.1 0.5 2.2

26 F 27 a cervix, radiation enterltln, bacterial overgrowth 3.9 1.9 2.1

27 60 H sprue, pernicious anemia 4.6 2.4 1.9

28 6B M sprue 0.9 0.5 1.8

29 52 F sprue 5.8 3.7 1.6

30 F mesenterlc. infarction, X sm. bowel resection 0.3 0.2 1.5

31 47 K B sprue 13.3 9.5 1.4

32 27 F pernicious anemia 18.4 13.9 1.3

33 72 F 9 sclcrodcrma, glαrdlasls, bacterial overgrowth 17.7 14.7 1.2

34 " 30 M Crolm'- disease, Heal resection __)_.__ J__6 Xd mean (6.7) (4.8) (1.6) range (0 .3-18.4) (0.2-14. 7) (1 .2-2.2)

Patlenta with pancreatic disease and other gastrointestinal diseases

35 57 chronic pancreatitis, pacudocyat, Crohn's dlscssc, Heal resection 2.3 0.4 5.8 36 45 19 pseudocyst, Vhlpplc procedure, 90! pancrcactectomy, sprue 4.3 0.2 21.5

Patients with pancreatic disease and renal Insu ficiency

37 54 F 40 chronic pancreatitis, Etoh, pancreatic calcification, renal 5.7 1.3 4.4 insufficiency (creatinine clearance 6 ml/mln)

38 61 F adenocarcinom-. pnncrcαs, whipplc procedure, total pancrcatectomy, 6.6 0.3 22.0 renal insufficiency (crentlnlnc clearance 20 ml/mln)

'Orally administered solutions were the same as those used in Table Itexcept -h» In 20 subjects 1. 5, 6, 10, 12, 16, 17 23, 27, 32, 31 and 35 the radioactive forms ofB__> were n ersed such that -Cojlljj; was given bound to IF end ("Ci-illjp w.i

_A plvcn bound to hop, R. SJKrn.i l <7 p/ilfiv

Table 5

Results of dual label pancreatic tests performed on normals, and patients with pancreatic and other gastrointestinal diseases

1 53 F None 24.9 1 133..BB 1 1..88

2 24 F 16.4 1 100..00 1 1..66

3 28 F 26.7 1 199..33 ^{• •} 11..44

4 35 M 21,0 1 166..88 1 1..33

5 62 M 15.4 - ^■ _ 1 122..11 1 1..33

6 3 F 25.6 2 200..11 1 1..33 ■■

7 31 H 23.2 1 199..88 1 1..22 a 29 M 18.9 1 166..44 1 1..22

. 26 F 23.0 1 188..88 1 1..22

10 32 H 22.9 2 200..55 1 1..11

11 40 M 23.1 2 233..11 1 1..00 mean (21.9) (17.3) (1.3)

Tange (15 .4-26.7) (10.0-23.1) (1.0-1.8) mean + 2 S.D. W .4-29.4) ( 9.0-25.6) (0.7-1.9)

Pa ltints with pnπcrcntlc dlscnse nn st qtorrhca

12 66 F 47 ndenocarclnom.i pancreas, Vhlpplc prncedurc, 1003! pancrcatectomy 0.5 46.2

13 54 153 chronic pancreatitis, idlσpαthlc, pancreatic calcification 0.6 39.3

14 46 M 39 chronic pancreatitis, Etoh, 90. pancrcatectomy 0.9 20.4

15 IB H 51 cystic flbronls 0.7 19.4

16 67 M 38 chronic pnncrcntltls, Etoh, lndurntcd pnncrens at laporotomy 1.1 17.0

17 M 12 chronic pancreatitis, Etoh, pancreatic calcification 1.2 11.7

18 33 H IB chronic pancreatitis, Typo IV hyperllpldemlα 1.0 11.6

19 52 H 27 chronic pancreatitis, Etoh, pancreatic calcification 1.6 10.8

20 58 H 27 chronic pancreatitis, Etoh, pancreatic calcification 0.9 10.7

21 26 F 19 gun shot wound, Vhlpple procedure, 75% pancrcatectomy 2.2 6.9

(16.5) (1.1) (19.4) range (9.B-23.6) (0 .5-2.2) (6.9-46.2)

to be in agreement with the severity of steatorrhea since the 5 patients with the highest ratios were the same 5 patients that also had the highest fecal fat contents.

While correlation was not perfect, this may have been ^" due ^' to the fact that the patients were not on diets of defined fat content while feces were being collected for the fecal fat tests.

Using the test of the present invention, two other patients with pancreatic disease that did not cause stea- torrhea had normal test results while a third patient with

of B]_2 to Bi2- These results suggest t at t e tests o t present invention provide similar, and possibly somewhat superior results to the 72-hour fecal fat analysis in term of its sensitivity in detecting patients with pancreatic exocrine insufficiency.

The specificity of the test. of the present invention was assessed by performing it on 10 patients with a variet of gastrointestinal diseases that included Crohn's disease, sprue, small bowel resection, scleroderma, bacterial over¬ growth, giardiasis and pernicious anemia. Vitamin B-^2 malabsorption is frequently present in many of these k k _fe_fe diseases and values for B-^ originating as 'IF-B^ ranged widely from about 0.3% to 18.4% (mean 6.7%) in these pa-

.. k tients. Values for B-j^ originating as - ^~ ~- ^~ -] 2 ^were reduced proportionately, however, and ranged from about 0.2% to 14.7% (mean 4.8%). This resulted in values for the ratio k k k

B12 to Bi2 that ranged from about 1.2 to 2.2 (mean 1.6), which were very similar to the ratios of normal subjects.

** * Values for the B^2 to Bj_2 ratio in these subjects with gastrointestinal disease did not overlap with those ob¬ tained .for patients with .pancreatic disease severe enough to cause steatorrhea. It is of interest to note the fact that 4 patients within the intestinal disease group had documented stea-

torrhea . Thi s demonstrates that steatorrhea itself is not k k responsible for elevated values for the ratios of B^_ ₂ ^to

- * ^B 12 -

One patient with both pancreatic disease and Crohn's 5. disease, and a second patient with both pancreatic disease and sprue, were studied to determine if the test could de¬ tect pancreatic-.exocrine insufficiency in the face of super¬ imposed intestinal disease. In the first patient, with k k k k

Crohn;s disease, the value for B-^ originating as IF-B- _j ^ 0 was reduced to 3.2%, but the value for B,2 originating as R-B12 was reduced even further to 0.4% thus giving an ele¬ vated value of 5.8 for the ratio of B ₁₂ ^{to B} 12* ^{In the} k k second patient , with sprue, the value for B- _j ^ was reduced k to 6.6% but the value for B- _j? ^was reduced disproportionate- ly to 0.3% thus giving an elevated value of 22.0 for the k k k ratio of B]_2 to B]_2. These results indicate that, the test of the present invention is capable of detecting pancreatic exocrine insufficiency even in the presence of superimposed intestinal disease. Two patients with combined pancreatic and renal dis¬ ease were studied, since it is known that the urinary ex¬ cretion of B- _j _2 is reduced in patients with renal disease. k k ~~ k k

Both patients had low values for B_2 originating as IF-B12, 5.7% and 6.6% respectively, but their values for B^ ₂ origin- ating as R-B]_2 were reduced disproportionately to 1.3% and

0.3% resp ly, their values for the ratio of 22.0 respectively, were both elevated. These results indicate that the test of the present invention can detect pancreatic exocrine insuffi- ciency in patients with superimposed renal disease pro¬ vided that such patients are not anuric.

Using the equivalent of the present invention, plasma levels of absorbed B-,2 ^can be monitored and are especially useful in patients who are anuric. Plasma levels could also be used in subjects without renal disease, as has been done with standard B-^ ₂ absorption tests, although here the

inj ection of unlabelled B ₁₂ would be omitted since such omission increases plasma levels of recently absorbed labelled B ₁₂ -

In most of the studies presented in Table 5, ^' was given bound to intrinsic factor and ( ⁵⁸ Co)B ₁₂ was give bound to hog R protein. However, the labelling isotopes were reversed, in at least several subjects in each major group and this did not appear to influence the results of the test. It has been determined that the time course of the k k k urinary excretion of B_2 and B^2 i not critical. The time k-V course of the urinary excretion of B^2 originating as IF-B^2 ^anα " B]_2 originating as R~Bi2 f° ^{r one} patient with pancreatic exocrine insufficiency and one normal subject are presented in Table 6. The data indicate that in both cases the ratio of B^-j ^{to B} 12 ^was highest in the 0 to 8 hou samples and lowest in the 16 to 24 hour samples. However, differences were slight, during these time periods, and this suggests that the additional time required for pan- creatic enzymes to act on B-^ ₂ originating as hog R-B^ ^anα ^ effect the transfer of such B12 ^{to IF} i ^s small compared with the time required for such B- _j _2 ^to reach the lleum and be absorbed. Because the ratio differences at various times were small, the inadvertent loss of individual urine samples during a 24-hour collection would not be expected to have a major effect on the value obtained for the ratio. The data presented in Table 6 also indicate that one could utilize 0 to 8 hour urine collections to evaluate patients for pancreatic exocrine insufficiency if time was a criti- cal factor.

The model upon which the present invention is based can also serve to explain certain anomalies noted in situa¬ tions of B- _] 2 absorption in certain cases of pancreatic exocrine dysfunction. Vitamin B-^ malabsorption in pan- creatic insufficiency has been corrected by the oral admin¬ istration of bicarbonate, pancreatic proteases or the B

Table 6

protein-( ^■:)0 Co)B*2 in pancreatic tests.

Percentage of administered B]_2

Time after excreted in 2- hour urine

Subject oral administration JF-Bf ^" R-3Ϊ2 IF- -3ιϊΗ-B| ₂

(Hr) (%) {%)

Patient with 0 - 8 8.52 0.45 18.9 pancreatic 8 - 16 5.02 0.27 18.6 exocrine 16 - 24 0.060 0.004 15.0 insufficiency

0 - 24 13.600 0.724 18.9

Normal ^" ' 0 - 8 4.30 3.53 1.22

8 - 16 .0.59 9.30 1.14

16 - 24 3.97 3.53 1.12

0 24 18.86 16.36 1.15

"Orally administered solution were the saine as those used in Table _. ⁺ The patient was No. 15 per Table -T. ^Subject No. 8 per Table -T .

analogue cobinamide. The model on which the present inven¬ tion is based provides an explanation for the action of these seemingly diverse B- _j ^ malabsorption correcting agents Bicarbonate exerts its effect by neutralizing gastric acid since at neutral (non-acid) pH IF can partially compete wit R protein for the initial binding and retention of B ₁₂ thu providing IF-B _] _2 for absorption. Pancreatic proteases func tion by partially degrading R protein and lowering its affinity for B-^ by 150-fold, and these changes result in the rapid transfer of B ₁₂ from R protein to IF, again pro¬ viding IF-B _] _2 for absorption. Cobinamide B,- analogue bind to R protein with high affinity, but does not bind to IF. This B^2 analogue functions by saturating endogenous R protein and leaving IF. as the only binder available for binding B- _j ^, thus again ^■ providing IF-B^ ₂ f° ^r absorption.

The model on which the present invention is based also provides an explanation for the fact that many patients wit severe pancreatic exocrine insufficiency have normal Schill ing tests since achlorhydria has the same effect as oral bicarbonate and is frequently superimposed with this exo-r crine insufficiency, and also since some variation does exist in the R protein content of basal gastric juice.

The data presented herein provides support for the present B _j _2 absorption model and the test of the present k invention. It shows that B^2 given orally bound to R pro¬ tein was malabsorbed markedly by all 10 subjects studied with pancreatic exocrine insufficiency severe enough to cause steatorrhea. Simultaneously administered B?ϊ gi ^ve n bound to IF, and given with cobinamide B^2 analogue resul- ted in the B ly normal a- ounts, by all of these same subjects. This observation provides strong evidence against the concept that pancrea¬ tic proteases alter the structure of IF in a manner that is required for B- _j ^ absorption. Preferential malabsorption k k of B^2 originating as -B^2 ^{was nofc} observed at all in a variety of intestinal diseases. This indicates that the

abnormalities in B _] _2 malabsorption frequently seen with intestinal diseases are not due to interference with the partial degradation of R protein by pancreatic proteases or with the transfer of B-, ₂ to intrinsic factor, even ^' though these steps do take place within the intestine.

The dual label test of the present invention is faster than the 72-hour fecal fat analysis and rs τnore convenient and less disagreeable for the subject and laboratory per¬ sonnel than either duodenal intubation tests or the fecal fat analysis. The present test also possesses the poten¬ tial of being much more easily available than either of the two current tests for pancreatic exocrine function since it could be performed by any physician with access to a rudimentary nuclear medicine facility, provided that the . test components were available.

The tests of the present invention are specific for pancreatic exocrine insufficiency. Such specificity is not possessed by the fecal fat analysis, since this test is also abnormal in a variety of intestinal diseases. Speci- ficity is important clinically since weight loss, diarrhea, and abdominal pain are common symptoms in both pancreatic and intestinal diseases. Specificity is also important in evaluating patients with known intestinal disease for the presence of superimposed pancreatic exocrine insuffi- ciency. k k originating as IF-B-,2 ^{to s tne} parameter which provides the test of the present invention with its specificity for - pancreatic exocrine insufficiency, it should be noted that useful information is also provided by the values obtained for itself since a marked reduction in was often observed in patients with intestinal disease. However, as marked reductions for B- _j ^.were ⁿ °t always observed, a normal k k value for IF-B^ ₂ does not rule out the presence of intes- tinal disease.

Although most of the studies presented here utilized

human IF, hog R protein, cobinamide ^~ -__2 ^a nalogue, and ( ⁵⁸ Co)B ₁₂ , ^ifc - ^s well established by the studies pre¬ sented here and elsewhere that similar results could be obtained by substitution of equivalent compositions for these components. For example, other species of IF, such as hog IF, rat IF, and others could..be used in place of or in combination with human IF. Other R proteins or other materials which bind B ₁₂ and make it unavailable for bind¬ ing to IF in the digestive system, unless pancreatic en- zymes are present, could be used with or in place of hog R protein. These include human R protein and dog R protein and any other material with these properties.

Several techniques, other than the use of B- _j ^ ana- logues exist for preventing endogenous R protein from re- moving B^2 from IF in the digestive system. One method would employ B^2 analogues in place of or in combination with cobinamide B- _j _2 analogue, which other B-^ analogues also possess the property of binding to human R protein with substantially the same affinity as B- _j _ ₂ while binding with low affinity to IF. B ₁₂ (bde-OH) and (3,4,6-Me ₃ benzimidazole)B-L2 ^are examples of some other B-j_2 analogues which possess these properties. Another method of prevent¬ ing R protein from removing from IF in the digestive system would provide for the effective removal or neutrali- zation of acid in the subject's stomach, since at neutral or slightly alkaline pH R protein is much less effective in removing B- _j _2 from IF than it i at acid pH. The acid could be neutralized by the oral administration of bicar¬ bonate or other antacids or by the oral or parenteral ad- ministration of agents, such as. the drug cimetidine, which limit the secretion of acid into the gastric juices. Many labelled forms of vitamin B _j2 are available for^use in the practice of the present invention, such as or ( ⁶⁰ Co)-B ₁₂ which could be utilized in place of ( ⁵⁷ Co)-B ₁ and ⁽⁵⁸ Co ⁾ -B ₁₂ *

Although the studies presented here measure the ab-

sorption of various labelled forms of B^ ₂ with the Schilling technique, it is well established that similar results could be obtained by using other methods of measuring the absorp¬ tion of labelled B ₁₂ . Examples of such applicable tech- niques include measuring labelled B-^ in feces or blood or by measuring the total body content or hepatic content of radioactive B]_2 with nuclear scanning.

Although the studies presented here utilize two differ¬ ent labelled forms of B-^ so that the absorption of B ₁₂ originating in both IF-B^2 ^{and R-B} 12 ^coul(3 e measured simultaneously, it is apparent that similar results could be obtained by using one labelled form of B-, ₂ ^a d measuring the absorption of IF-B^2 ^an< 3 ^R_B 12 ^on separate occasions. While preferred embodiments have set forth concentra- tions and amounts of various components utilized in certain studies, the useful and preferred amounts of those compo¬ nents are set forth below:

Composition Useful Range Preferred Range

IF-B ₁ 0.002-20.0 nmol 0.05-2.0 nmol -Bl2 0.002-20.0 nmol 0.05-2.0 nmol

Free IF 0.05-100.0 nmol 0.2-1.0 nmol

B ₁₂ analogues 0.2-20,000 nmol 20-500 nmol

While the invention has been particularly shown and described with reference to preferred embodiments thereof, it will be understood by those skilled in the art that the foregoing and other modifications or changes in form and detail may be made therein without departing from the spirit and scope of the invention. What is- claimed is: