Case Studies Drug-Induced Autoimmune Hemolytic Anemia in a 78-Year-Old African-American Man with Chronic Lymphocytic Leukemia Alex Sartori, MLS(ASCP)CM,1 Brian Staley, MT(ASCP),2 Angela Skipper, MLS(ASCP)CMSBBCM 3 Lab Med Summer 2014;45:e105-e108
CLINICAL HISTORY
Patient: 78-year-old African-American man. Past Medical History: Chronic lymphocytic leukemia first diagnosed in 2003, with a subsequent relapse in 2006 and another in 2010. History of Present Illness: In late 2011, the patient was admitted to the hospital for cholelithiasis, at which time his treating physician incidentally discovered severe anemia. The anemia worsened as time went on, and the patient became transfusion dependent. Hypogammaglobulinemia secondary to chronic lymphocytic leukemia (CLL) required that the patient receive intravenous immunoglobulin.
Abbreviations CLL, chronic lymphocytic leukemia; R-CHOP, rituximab, cyclophosphamide, hydroxydaunorubicin, oncovin, and prednisone; FCR, fludarabine, cyclophosphamide, and rituximab; IVIG, intravenous immunoglobulin; AHA, autoimmune hemolytic anemia; RBCs, red blood cells; D, Rh(D) antigen; C, Rh(C) antigen; E, Rh(E) antigen; c, Rh(c) antigen; e, Rh(e) antigen; f, Rh(f) antigen; Cw, Rh(Cw) antigen; V, Rh(V) antigen; K, Kell antigen; k, cellano antigen; Kpa, Kell A antigen; Kpb, Kell B antigen; Jsa, Kell system; Jsb, Kell system; Fya, Duffy A antigen; Fyb, Duffy B antigen; Jka, Kidd A antigen; Jk b, Kidd B antigen; Xga, Sex-linked antigen; Lea, Lewis A antigen; Leb, Lewis B antigen; S, MNS blood group system antigen; s, MNS blood group system antigen; M, MNS blood group system antigen; N, MNS blood group system antigen; P1, P blood group system; Lua, Lutheran A antigen; Lub, Lutheran B antigen; PT, Patient; DAT, direct antiglobulin testing; AHG, antihuman globulin; IgG, immunoglobulin G; IgM, immunoglobulin M; IS, Immediate Spin; CC, Check Cells; NA, not applicable; Int, Interpretation; Pos, positive; Neg, negative; SC, Screening Cells; A/C, Auto Control; Oi, Cord Blood sample (i pos). 1 Medical Laboratory Sciences Program, University of North Florida, Jacksonville, FL, 2Southeastern division, Immucor, Inc., Norcross, GA, 3 Mayo Clinic, Jacksonville, FL
*To whom correspondence should be addressed: E-mail: [email protected]
www.labmedicine.com
Despite transfusion therapy, the anemia failed to lessen; laboratory results eventually led to the diagnosis of a drug-induced warm autoantibody that triggered hemolytic anemia. Medications: The patient had taken rituximab in 2003; rituximab, cyclophosphamide, hydroxydaunorubicin, oncovin, and prednisone (R-CHOP) in 2006; fludarabine, cyclophosphamide, and rituximab (FCR) in 2010; and intravenous Immunoglobulin (IVIG) and prednisone in 2011. Keywords: chronic lymphocytic leukemia, hemolytic anemia, drug induced, warm autoantibody, autoimmune, transfusion
Questions 1. What laboratory results lead to the diagnosis of warm autoantibody? 2. How did the results rule out other possible causes in this case? 3. What are some possible etiologies of autoantibodies? 4. How can drug-induced autoantibodies form? 5. What are some drugs that are known to induce antibody formation? 6. How can drug-induced autoimmune hemolytic anemia (AHA) be differentiated from warm AHA? 7. What are some approaches to treating a patient with warm AHA? 1. Before transfusion, the immediate spin crossmatch for the patient was incompatible with every ABO-Rh compatible unit tested. The initial antibody screen results for the patient were 3+ positive for screening cells I and II. We performed a full antibody identification panel; the specimen from our patient was 2+/3+ reactive with all cells tested, including an auto control with serum/red
Summer 2014 | Volume 45, Number 3 Lab Medicine e105
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
DOI: 10.1309/LMLWQDB2Q6LS7VQG
Case Studies
blood cells (RBCs) from the patient (Table 1). Results of direct antiglobulin testing (DAT) were 3+ reactive with polyspecific antihuman globulin (AHG) reagent, as well as with monospecific antiimmunoglobulin G (IgG). However, the patient’s serum was nonreactive with anti-C3d and the saline control (Table 2). We performed a mini cold panel, with the patient specimen being reactive with all cells at 4°C but not at 37°C (Table 3). We performed an acid elution, with the eluate continuing to be reactive with all panel cells (Table 4).
3. Due to the nature of these diseases, lymphoproliferative disorders generally confer a high rate of autoimmune complications. The most common complications observed in CLL are AHA, idiopathic thrombocytopenia purpura, and pure RBC aplasia.1 Of these, AHA is the most common type of complication that
4. Drug-induced autoantibodies can be broadly divided into 2 main categories: drug-dependent and drugindependent. Drug-dependent autoantibodies are the most common; they require the presence of the drug in vitro to detect the antibody. Drug-independent antibodies are less common; however, scientists can detect them in vitro without the addition of the drug. Although the exact mechanisms of in vivo hemolysis are not fully understood, there are a few possible explanations. Regarding drug-dependent antibody formation, the most accepted hypothesis involves covalent binding of the drug to erythrocyte membrane proteins. An antibody (usually IgG) can be created against the drug, which then binds to the drug-coated erythrocyte and is subsequently cleared
Table 1. Gel Antibody Panel Results Showing Panagglutination with and Without Autocontrol No. D C E c e f
Cw V K k Kpa Kpb Jsa Jsb Fya Fyb Jka Jkb Xga Lea Leb S s M N P1 Lua Lub Result
1 + + 0 0 + 0 + 0 0 + 0 + 0 + + + 0 + + 0 + + + + 0 + 0 + 2+ 2 + + 0 0 + 0 0 0 0 + 0 + / + + + + 0 + + 0 + + + 0 + 0 + 2+ 3 + + + + 0 0 0 0 0 + 0 + / + + 0 + + + 0 + + + + 0 + 0 + 2+ 4 + 0 0 + + + 0 0 0 + 0 + 0 + 0 0 + + 0 0 0 0 + + 0 + 0 + 2+ 5 0 0 0 + + + 0 0 0 + 0 + / + 0 + + + 0 0 0 + 0 + 0 + 0 + 2+ 6 0 0 + + + + 0 0 0 + 0 + / + 0 + + + 0 0 + 0 + 0 + + 0 + 2+ 7 0 0 0 + + + 0 0 + + 0 + / + 0 + + + 0 0 + + + + + 0 0 + 3+ 8 0 0 0 + + + 0 0 0 + 0 + / + + 0 + 0 + + 0 0 + + 0 0 0 + 3+ 9 0 0 0 + + + 0 0 0 + 0 + 0 + + + 0 + + 0 + + 0 + 0 + 0 + 3+ 10 0 0 0 + + + 0 0 0 + 0 + 0 + 0 + + + + + 0 + + 0 + + 0 + 3+ 11 0 + 0 0 0 0 0 0 + + 0 + / + 0 + + + + 0 + + + + + + 0 + 3+ PT 3+ D, Rh(D) antigen; C, Rh(C) antigen; E, Rh(E) antigen; c, Rh(c) antigen; e, Rh(e) antigen; f, Rh(f) antigen; Cw, Rh(Cw) antigen; V, Rh(V) antigen ; K, Kell antigen; k, cellano antigen; Kpa, Kell A antigen; Kpb, Kell B antigen; Jsa, Kell system; Jsb, Kell system; Fya, Duffy A antigen Fyb, Duffy B antigen; Jka, Kidd A antigen; Jkb, Kidd B antigen; Xga, Sex-linked antigen; Lea, Lewis A antigen; Leb, Lewis B antigen; S, MNS blood group system antigen; s, MNS blood group system antigen; M, MNS blood group system antigen; N, MNS blood group system antigen; P1, P blood group system; Lua, Lutheran A antigen; Lub, Lutheran B antigen; PT, Patient
e106
Lab Medicine Spring 2014 | Volume 45, Number 2
www.labmedicine.com
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
2. We ruled out other causes of panagglutination through the testing process. We ruled out anti-I due to the fact that the patient specimen had reacted with all 3 cord blood samples, which contain the i antigen but no I antigen. We ruled out an antibody to a high-frequency antigen such as k (cellano) because the specimen was antigen typed; we ordered an autoadsorption process to be performed by a local reference laboratory (The Blood Alliance, Jacksonville, FL). We ruled out cold agglutinins because the patient specimen was reactive with prewarmed cells and cells at the Coombs phase of testing. The fact that the autocontrol material tested negative provided further evidence against an alloantibody as the culprit.
occurs in CLL, appearing in as many as 25% of all cases. 2 The exact mechanism of CLL-induced AHA is unknown; theories range from failure of T-helper regulation2 to leukemic cells acting as antigen-presenting cells, which expose part of the Rh epitope that had been previously concealed.3 Another possible explanation for the AHA in our patient is that it was caused by a chemotherapy drug or even the intravenous immunoglobulin that the patient had been receiving; intravenous immunoglobulin has been associated in the literature with hemolytic anemia.4 In our patient, his anemia began shortly after he had started chemotherapy to treat his second relapse; this suggests that the chemotherapy had had some effect on his treatment course. The most likely scenario, however, is that the patient had underlying, less severe AHA secondary to CLL, which was exacerbated by the treatment.
Case Studies
Table 2. Results of DAT on a Specimen from the Patienta
Table 3. Results of a Mini Cold Panel on the Serum of the Patienta
Variable Polyspecific Anti-IgG
Anti-C3d Saline
Selected Cells
4°C
0 0 1+ Neg
SC I SC II A/C Oi I Oi II Oi III
3+ 0 3+ 0 3+ 0 3+ 0 3+ 0 3+ 0
IS 3+ 5 min 3+ CC NA Int Pos
3+ 3+ NA Pos
0 0 NA Neg
DAT, direct antiglobulin testing; IgG, immunoglobulin G; IS, Immediate Spin; CC,Check Cells; NA, not applicable; Int, Interpretation; Pos, positive; Neg, negative. a Patient is a 78-year-old African-American man.
5. There are several drugs known to induce antibody formation. Drug-dependent antibody formation is most commonly caused by penicillin and piperacillin, along with cefotetan and ceftriaxone. Historically, drug-independent antibodies were formed most commonly in response to methyldopa, a drug used to treat hypertension. Today, purine analogs (such as fludarabine) used in the production of chemotherapy drugs have become recognized as inducers of antibody formation; however, the overall incidence of this occurrence remains unclear. This is due
SC, Screening Cells; A/C, Auto Control; Oi, Cord Blood Sample. a Patient is a 78-year-old African-American man.
mostly to the fact that many of the patients receiving these drugs already have underlying autoimmune complications, which result in difficulties in measuring the occurrence of antibody formation occurring strictly as a result of the drugs themselves. However, many scientists believe colloquially that overall, the incidence of autoimmune complications from purine analogs is relatively rare but can be severe. In modern times, the most common drug known to cause drug-independent antibody formation is fludarabine,6 which our patient had been taking after his second relapse. 6. Drug-independent antibody formation mimics a true autoantibody, even without the in vitro presence of the drug; therefore, distinguishing the drug-dependent and drug-independent versions is difficult to impossible. The only real means of differentiation is cessation of ingesting the drug in question. In patients with underlying AHA secondary to CLL, even drug cessation may
Table 4. Results of an Antibody Panel Performed on Eluatea No. D C E c e f
Cw V K k Kpa Kpb Jsa Jsb Fya Fyb Jka Jkb Xga Lea Leb S s M N P1 Lua Lub Result
1 + + 0 0 + 0 + 0 0 + 0 + 0 + + + 0 + + 0 + + + + 0 + 0 + 2+ 2 + + 0 0 + 0 0 0 0 + 0 + / + + + + 0 + + 0 + + + 0 + 0 + 2+ 3 + 0 + + 0 0 0 0 0 + 0 + / + + 0 + + + 0 + + + + 0 + 0 + 2+ 4 + 0 0 + + + 0 0 0 + 0 + 0 + 0 0 + + 0 0 0 0 + + 0 + 0 + 2+ 5 0 0 0 + + + 0 0 0 + 0 + / + 0 + + + 0 0 0 + 0 + 0 + 0 + 2+ 6 0 0 + + + + 0 0 0 + 0 + / + 0 + + + 0 0 + 0 + 0 + + 0 + 2+ 7 0 0 0 + + + 0 0 + + 0 + / + 0 + + + 0 0 + + + + + 0 0 + 3+ 8 0 0 0 + + + 0 0 0 + 0 + / + + 0 + 0 + + 0 0 + + 0 0 0 + 3+ 9 0 0 0 + + + 0 0 0 + 0 + 0 + + + 0 + + 0 + + 0 + 0 + 0 + 3+ 10 0 0 0 + + + 0 0 0 + 0 + 0 + 0 + + + + + 0 + + 0 + + 0 + 3+ 11 0 + 0 0 0 0 0 0 + + 0 + / + 0 + + + + 0 + + + + + + 0 + 3+ D, Rh(D) antigen; C, Rh(C) antigen; E, Rh(E) antigen; c, Rh(c) antigen; e, Rh(e) antigen; f, Rh(f) antigen; Cw, Rh(Cw) antigen; V, Rh(V) antigen ; K, Kell antigen; k, cellano antigen; Kpa, Kell A antigen; Kpb, Kell B antigen; Jsa, Kell system; Jsb, Kell system; Fya, Duffy A antigen Fyb, Duffy B antigen; Jka, Kidd A antigen; Jkb, Kidd B antigen; Xga, Sex-linked antigen; Lea, Lewis A antigen; Leb, Lewis B antigen; S, MNS blood group system antigen; s, MNS blood group system antigen; M, MNS blood group system antigen; N, MNS blood group system antigen; P1, P blood group system; Lua, Lutheran A antigen; Lub, Lutheran B antigen; PT, Patient a Initial specimen material provided by the patient, a 78-year-old African-American man.
www.labmedicine.com
Spring 2014 | Volume 45, Number 2 Lab Medicine e107
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
by the macrophages in the spleen. Immunoglobulin M (IgM) antibodies form less commonly; when they are created, they can activate complement and cause acute intravascular hemolysis. Drug-independent antibody formation is less understood than the drug-dependent type and may possibly involve drug adsorption to the erythrocyte membrane, creating a neoantigen composed of membrane and drug.5
37°C
Case Studies
The patient has been receiving transfusions only when necessary, namely, when his hemoglobin level drops below 7 g/dL. However, CLL with autoimmune complications carries an overall poor prognosis. To treat these complications, the patient has been receiving prednisone and IVIG. LM
7. If the hemolytic anemia is drug induced, cessation of ingesting the causative drug can correct the anemia, as long as no other underlying causes of anemia are present. However, with CLL, there is often underlying AHA that is exacerbated by a drug-induced autoantibody. If that is so, drug cessation will slow but not stop the hemolytic process. Transfusing a patient with a warm autoantibody can greatly accelerate hemolysis; thus, we do not recommend this practice. However, this is sometimes unavoidable. As an added complication, nonspecific binding of the autoantibody can mask the detection of clinically significant alloantibodies and can cause DATpositive RBCs to be removed in the spleen. If transfusion is necessary, adsorption studies can be performed to detect any underlying clinically significant alloantibodies. Antigen typing of the RBCs of the patients and providing matched units is also possible. Two drug treatments include corticosteroids and IVIG. Corticosteroids are used to reduce the clearing of antibody-coated RBCs, and IVIG can aid in this process by inundating the immune system, further slowing the hemolytic process.
Acknowledgements
e108
Lab Medicine Spring 2014 | Volume 45, Number 2
The primary author thanks his coauthors, whose combined mentorship, knowledge, expertise, and patience have been crucial to the preparation of this case report.
References 1. Mauro FR, Foa R, Cerretti R, et al. Autoimmune hemolytic anemia in chronic lymphocytic leukemia: clinical, therapeutic, and prognostic features. Blood. 2000;95(9):2786-2792. 2. Hamblin T. Autoimmune complications in chronic lymphocytic leukemia. Semin Oncol. 2006;33(2):230-239. 3. D’Arena G, Cascavilla N. Chronic lymphocytic leukemiaassociated autoimmune hemolytic anemia. Leuk Lymphoma. 2007;48(6): 1072-1080. 4. Wilson JR, Bhoopalam H, Fisher M. Hemolytic anemia associated with intravenous immunoglobulin. Muscle Nerve. 1997;20(9):1142-1145. 5. Garratty G. 2010. Immune hemolytic anemia associated with drug therapy. Blood Rev. 2010;24:143-150. 6. Pierce A, Nester R, for the Education Committee of the Academy of Clinical Laboratory Physicians and Scientists. Pathology consultation on drug-induced hemolytic anemia. Am J Clin Pathol. 2011;136(1):7-12.
www.labmedicine.com
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
not definitively differentiate the 2 types. Regarding distinguishing between drug-independent and drugdependent antibody formation, scientists can examine the reactivity of the eluate from an acid elution. With drug-dependent antibody formation, the eluate will be characteristically nonreactive due to the absence of the drug, whereas a drug-independent antibody will still be reactive.
CLINICAL HISTORY
Patient: 78-year-old African-American man. Past Medical History: Chronic lymphocytic leukemia first diagnosed in 2003, with a subsequent relapse in 2006 and another in 2010. History of Present Illness: In late 2011, the patient was admitted to the hospital for cholelithiasis, at which time his treating physician incidentally discovered severe anemia. The anemia worsened as time went on, and the patient became transfusion dependent. Hypogammaglobulinemia secondary to chronic lymphocytic leukemia (CLL) required that the patient receive intravenous immunoglobulin.
Abbreviations CLL, chronic lymphocytic leukemia; R-CHOP, rituximab, cyclophosphamide, hydroxydaunorubicin, oncovin, and prednisone; FCR, fludarabine, cyclophosphamide, and rituximab; IVIG, intravenous immunoglobulin; AHA, autoimmune hemolytic anemia; RBCs, red blood cells; D, Rh(D) antigen; C, Rh(C) antigen; E, Rh(E) antigen; c, Rh(c) antigen; e, Rh(e) antigen; f, Rh(f) antigen; Cw, Rh(Cw) antigen; V, Rh(V) antigen; K, Kell antigen; k, cellano antigen; Kpa, Kell A antigen; Kpb, Kell B antigen; Jsa, Kell system; Jsb, Kell system; Fya, Duffy A antigen; Fyb, Duffy B antigen; Jka, Kidd A antigen; Jk b, Kidd B antigen; Xga, Sex-linked antigen; Lea, Lewis A antigen; Leb, Lewis B antigen; S, MNS blood group system antigen; s, MNS blood group system antigen; M, MNS blood group system antigen; N, MNS blood group system antigen; P1, P blood group system; Lua, Lutheran A antigen; Lub, Lutheran B antigen; PT, Patient; DAT, direct antiglobulin testing; AHG, antihuman globulin; IgG, immunoglobulin G; IgM, immunoglobulin M; IS, Immediate Spin; CC, Check Cells; NA, not applicable; Int, Interpretation; Pos, positive; Neg, negative; SC, Screening Cells; A/C, Auto Control; Oi, Cord Blood sample (i pos). 1 Medical Laboratory Sciences Program, University of North Florida, Jacksonville, FL, 2Southeastern division, Immucor, Inc., Norcross, GA, 3 Mayo Clinic, Jacksonville, FL
*To whom correspondence should be addressed: E-mail: [email protected]
www.labmedicine.com
Despite transfusion therapy, the anemia failed to lessen; laboratory results eventually led to the diagnosis of a drug-induced warm autoantibody that triggered hemolytic anemia. Medications: The patient had taken rituximab in 2003; rituximab, cyclophosphamide, hydroxydaunorubicin, oncovin, and prednisone (R-CHOP) in 2006; fludarabine, cyclophosphamide, and rituximab (FCR) in 2010; and intravenous Immunoglobulin (IVIG) and prednisone in 2011. Keywords: chronic lymphocytic leukemia, hemolytic anemia, drug induced, warm autoantibody, autoimmune, transfusion
Questions 1. What laboratory results lead to the diagnosis of warm autoantibody? 2. How did the results rule out other possible causes in this case? 3. What are some possible etiologies of autoantibodies? 4. How can drug-induced autoantibodies form? 5. What are some drugs that are known to induce antibody formation? 6. How can drug-induced autoimmune hemolytic anemia (AHA) be differentiated from warm AHA? 7. What are some approaches to treating a patient with warm AHA? 1. Before transfusion, the immediate spin crossmatch for the patient was incompatible with every ABO-Rh compatible unit tested. The initial antibody screen results for the patient were 3+ positive for screening cells I and II. We performed a full antibody identification panel; the specimen from our patient was 2+/3+ reactive with all cells tested, including an auto control with serum/red
Summer 2014 | Volume 45, Number 3 Lab Medicine e105
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
DOI: 10.1309/LMLWQDB2Q6LS7VQG
Case Studies
blood cells (RBCs) from the patient (Table 1). Results of direct antiglobulin testing (DAT) were 3+ reactive with polyspecific antihuman globulin (AHG) reagent, as well as with monospecific antiimmunoglobulin G (IgG). However, the patient’s serum was nonreactive with anti-C3d and the saline control (Table 2). We performed a mini cold panel, with the patient specimen being reactive with all cells at 4°C but not at 37°C (Table 3). We performed an acid elution, with the eluate continuing to be reactive with all panel cells (Table 4).
3. Due to the nature of these diseases, lymphoproliferative disorders generally confer a high rate of autoimmune complications. The most common complications observed in CLL are AHA, idiopathic thrombocytopenia purpura, and pure RBC aplasia.1 Of these, AHA is the most common type of complication that
4. Drug-induced autoantibodies can be broadly divided into 2 main categories: drug-dependent and drugindependent. Drug-dependent autoantibodies are the most common; they require the presence of the drug in vitro to detect the antibody. Drug-independent antibodies are less common; however, scientists can detect them in vitro without the addition of the drug. Although the exact mechanisms of in vivo hemolysis are not fully understood, there are a few possible explanations. Regarding drug-dependent antibody formation, the most accepted hypothesis involves covalent binding of the drug to erythrocyte membrane proteins. An antibody (usually IgG) can be created against the drug, which then binds to the drug-coated erythrocyte and is subsequently cleared
Table 1. Gel Antibody Panel Results Showing Panagglutination with and Without Autocontrol No. D C E c e f
Cw V K k Kpa Kpb Jsa Jsb Fya Fyb Jka Jkb Xga Lea Leb S s M N P1 Lua Lub Result
1 + + 0 0 + 0 + 0 0 + 0 + 0 + + + 0 + + 0 + + + + 0 + 0 + 2+ 2 + + 0 0 + 0 0 0 0 + 0 + / + + + + 0 + + 0 + + + 0 + 0 + 2+ 3 + + + + 0 0 0 0 0 + 0 + / + + 0 + + + 0 + + + + 0 + 0 + 2+ 4 + 0 0 + + + 0 0 0 + 0 + 0 + 0 0 + + 0 0 0 0 + + 0 + 0 + 2+ 5 0 0 0 + + + 0 0 0 + 0 + / + 0 + + + 0 0 0 + 0 + 0 + 0 + 2+ 6 0 0 + + + + 0 0 0 + 0 + / + 0 + + + 0 0 + 0 + 0 + + 0 + 2+ 7 0 0 0 + + + 0 0 + + 0 + / + 0 + + + 0 0 + + + + + 0 0 + 3+ 8 0 0 0 + + + 0 0 0 + 0 + / + + 0 + 0 + + 0 0 + + 0 0 0 + 3+ 9 0 0 0 + + + 0 0 0 + 0 + 0 + + + 0 + + 0 + + 0 + 0 + 0 + 3+ 10 0 0 0 + + + 0 0 0 + 0 + 0 + 0 + + + + + 0 + + 0 + + 0 + 3+ 11 0 + 0 0 0 0 0 0 + + 0 + / + 0 + + + + 0 + + + + + + 0 + 3+ PT 3+ D, Rh(D) antigen; C, Rh(C) antigen; E, Rh(E) antigen; c, Rh(c) antigen; e, Rh(e) antigen; f, Rh(f) antigen; Cw, Rh(Cw) antigen; V, Rh(V) antigen ; K, Kell antigen; k, cellano antigen; Kpa, Kell A antigen; Kpb, Kell B antigen; Jsa, Kell system; Jsb, Kell system; Fya, Duffy A antigen Fyb, Duffy B antigen; Jka, Kidd A antigen; Jkb, Kidd B antigen; Xga, Sex-linked antigen; Lea, Lewis A antigen; Leb, Lewis B antigen; S, MNS blood group system antigen; s, MNS blood group system antigen; M, MNS blood group system antigen; N, MNS blood group system antigen; P1, P blood group system; Lua, Lutheran A antigen; Lub, Lutheran B antigen; PT, Patient
e106
Lab Medicine Spring 2014 | Volume 45, Number 2
www.labmedicine.com
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
2. We ruled out other causes of panagglutination through the testing process. We ruled out anti-I due to the fact that the patient specimen had reacted with all 3 cord blood samples, which contain the i antigen but no I antigen. We ruled out an antibody to a high-frequency antigen such as k (cellano) because the specimen was antigen typed; we ordered an autoadsorption process to be performed by a local reference laboratory (The Blood Alliance, Jacksonville, FL). We ruled out cold agglutinins because the patient specimen was reactive with prewarmed cells and cells at the Coombs phase of testing. The fact that the autocontrol material tested negative provided further evidence against an alloantibody as the culprit.
occurs in CLL, appearing in as many as 25% of all cases. 2 The exact mechanism of CLL-induced AHA is unknown; theories range from failure of T-helper regulation2 to leukemic cells acting as antigen-presenting cells, which expose part of the Rh epitope that had been previously concealed.3 Another possible explanation for the AHA in our patient is that it was caused by a chemotherapy drug or even the intravenous immunoglobulin that the patient had been receiving; intravenous immunoglobulin has been associated in the literature with hemolytic anemia.4 In our patient, his anemia began shortly after he had started chemotherapy to treat his second relapse; this suggests that the chemotherapy had had some effect on his treatment course. The most likely scenario, however, is that the patient had underlying, less severe AHA secondary to CLL, which was exacerbated by the treatment.
Case Studies
Table 2. Results of DAT on a Specimen from the Patienta
Table 3. Results of a Mini Cold Panel on the Serum of the Patienta
Variable Polyspecific Anti-IgG
Anti-C3d Saline
Selected Cells
4°C
0 0 1+ Neg
SC I SC II A/C Oi I Oi II Oi III
3+ 0 3+ 0 3+ 0 3+ 0 3+ 0 3+ 0
IS 3+ 5 min 3+ CC NA Int Pos
3+ 3+ NA Pos
0 0 NA Neg
DAT, direct antiglobulin testing; IgG, immunoglobulin G; IS, Immediate Spin; CC,Check Cells; NA, not applicable; Int, Interpretation; Pos, positive; Neg, negative. a Patient is a 78-year-old African-American man.
5. There are several drugs known to induce antibody formation. Drug-dependent antibody formation is most commonly caused by penicillin and piperacillin, along with cefotetan and ceftriaxone. Historically, drug-independent antibodies were formed most commonly in response to methyldopa, a drug used to treat hypertension. Today, purine analogs (such as fludarabine) used in the production of chemotherapy drugs have become recognized as inducers of antibody formation; however, the overall incidence of this occurrence remains unclear. This is due
SC, Screening Cells; A/C, Auto Control; Oi, Cord Blood Sample. a Patient is a 78-year-old African-American man.
mostly to the fact that many of the patients receiving these drugs already have underlying autoimmune complications, which result in difficulties in measuring the occurrence of antibody formation occurring strictly as a result of the drugs themselves. However, many scientists believe colloquially that overall, the incidence of autoimmune complications from purine analogs is relatively rare but can be severe. In modern times, the most common drug known to cause drug-independent antibody formation is fludarabine,6 which our patient had been taking after his second relapse. 6. Drug-independent antibody formation mimics a true autoantibody, even without the in vitro presence of the drug; therefore, distinguishing the drug-dependent and drug-independent versions is difficult to impossible. The only real means of differentiation is cessation of ingesting the drug in question. In patients with underlying AHA secondary to CLL, even drug cessation may
Table 4. Results of an Antibody Panel Performed on Eluatea No. D C E c e f
Cw V K k Kpa Kpb Jsa Jsb Fya Fyb Jka Jkb Xga Lea Leb S s M N P1 Lua Lub Result
1 + + 0 0 + 0 + 0 0 + 0 + 0 + + + 0 + + 0 + + + + 0 + 0 + 2+ 2 + + 0 0 + 0 0 0 0 + 0 + / + + + + 0 + + 0 + + + 0 + 0 + 2+ 3 + 0 + + 0 0 0 0 0 + 0 + / + + 0 + + + 0 + + + + 0 + 0 + 2+ 4 + 0 0 + + + 0 0 0 + 0 + 0 + 0 0 + + 0 0 0 0 + + 0 + 0 + 2+ 5 0 0 0 + + + 0 0 0 + 0 + / + 0 + + + 0 0 0 + 0 + 0 + 0 + 2+ 6 0 0 + + + + 0 0 0 + 0 + / + 0 + + + 0 0 + 0 + 0 + + 0 + 2+ 7 0 0 0 + + + 0 0 + + 0 + / + 0 + + + 0 0 + + + + + 0 0 + 3+ 8 0 0 0 + + + 0 0 0 + 0 + / + + 0 + 0 + + 0 0 + + 0 0 0 + 3+ 9 0 0 0 + + + 0 0 0 + 0 + 0 + + + 0 + + 0 + + 0 + 0 + 0 + 3+ 10 0 0 0 + + + 0 0 0 + 0 + 0 + 0 + + + + + 0 + + 0 + + 0 + 3+ 11 0 + 0 0 0 0 0 0 + + 0 + / + 0 + + + + 0 + + + + + + 0 + 3+ D, Rh(D) antigen; C, Rh(C) antigen; E, Rh(E) antigen; c, Rh(c) antigen; e, Rh(e) antigen; f, Rh(f) antigen; Cw, Rh(Cw) antigen; V, Rh(V) antigen ; K, Kell antigen; k, cellano antigen; Kpa, Kell A antigen; Kpb, Kell B antigen; Jsa, Kell system; Jsb, Kell system; Fya, Duffy A antigen Fyb, Duffy B antigen; Jka, Kidd A antigen; Jkb, Kidd B antigen; Xga, Sex-linked antigen; Lea, Lewis A antigen; Leb, Lewis B antigen; S, MNS blood group system antigen; s, MNS blood group system antigen; M, MNS blood group system antigen; N, MNS blood group system antigen; P1, P blood group system; Lua, Lutheran A antigen; Lub, Lutheran B antigen; PT, Patient a Initial specimen material provided by the patient, a 78-year-old African-American man.
www.labmedicine.com
Spring 2014 | Volume 45, Number 2 Lab Medicine e107
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
by the macrophages in the spleen. Immunoglobulin M (IgM) antibodies form less commonly; when they are created, they can activate complement and cause acute intravascular hemolysis. Drug-independent antibody formation is less understood than the drug-dependent type and may possibly involve drug adsorption to the erythrocyte membrane, creating a neoantigen composed of membrane and drug.5
37°C
Case Studies
The patient has been receiving transfusions only when necessary, namely, when his hemoglobin level drops below 7 g/dL. However, CLL with autoimmune complications carries an overall poor prognosis. To treat these complications, the patient has been receiving prednisone and IVIG. LM
7. If the hemolytic anemia is drug induced, cessation of ingesting the causative drug can correct the anemia, as long as no other underlying causes of anemia are present. However, with CLL, there is often underlying AHA that is exacerbated by a drug-induced autoantibody. If that is so, drug cessation will slow but not stop the hemolytic process. Transfusing a patient with a warm autoantibody can greatly accelerate hemolysis; thus, we do not recommend this practice. However, this is sometimes unavoidable. As an added complication, nonspecific binding of the autoantibody can mask the detection of clinically significant alloantibodies and can cause DATpositive RBCs to be removed in the spleen. If transfusion is necessary, adsorption studies can be performed to detect any underlying clinically significant alloantibodies. Antigen typing of the RBCs of the patients and providing matched units is also possible. Two drug treatments include corticosteroids and IVIG. Corticosteroids are used to reduce the clearing of antibody-coated RBCs, and IVIG can aid in this process by inundating the immune system, further slowing the hemolytic process.
Acknowledgements
e108
Lab Medicine Spring 2014 | Volume 45, Number 2
The primary author thanks his coauthors, whose combined mentorship, knowledge, expertise, and patience have been crucial to the preparation of this case report.
References 1. Mauro FR, Foa R, Cerretti R, et al. Autoimmune hemolytic anemia in chronic lymphocytic leukemia: clinical, therapeutic, and prognostic features. Blood. 2000;95(9):2786-2792. 2. Hamblin T. Autoimmune complications in chronic lymphocytic leukemia. Semin Oncol. 2006;33(2):230-239. 3. D’Arena G, Cascavilla N. Chronic lymphocytic leukemiaassociated autoimmune hemolytic anemia. Leuk Lymphoma. 2007;48(6): 1072-1080. 4. Wilson JR, Bhoopalam H, Fisher M. Hemolytic anemia associated with intravenous immunoglobulin. Muscle Nerve. 1997;20(9):1142-1145. 5. Garratty G. 2010. Immune hemolytic anemia associated with drug therapy. Blood Rev. 2010;24:143-150. 6. Pierce A, Nester R, for the Education Committee of the Academy of Clinical Laboratory Physicians and Scientists. Pathology consultation on drug-induced hemolytic anemia. Am J Clin Pathol. 2011;136(1):7-12.
www.labmedicine.com
Downloaded from http://labmed.oxfordjournals.org/ by guest on June 5, 2016
not definitively differentiate the 2 types. Regarding distinguishing between drug-independent and drugdependent antibody formation, scientists can examine the reactivity of the eluate from an acid elution. With drug-dependent antibody formation, the eluate will be characteristically nonreactive due to the absence of the drug, whereas a drug-independent antibody will still be reactive.
