International Journal of Laboratory Hematology The Official journal of the International Society for Laboratory Hematology
ORIGINAL ARTICLE
INTERNAT IONAL JOURNAL OF LABORATO RY HEMATO LOGY
A quality improvement assessment of multiple, concurrent flow cytometry analyses at a tertiary care center H. E. KARNES, J. L. FRATER
Department of Pathology & Immunology, Washington University in St. Louis School of Medicine, St. Louis, MO, USA Correspondence: John L. Frater, Department of Pathology & Immunology, Washington University in St. Louis School of Medicine, 660 S. Euclid Avenue, Campus Box 8118, St. Louis, MO 63110, USA. Tel.: (314) 362-1553; Fax: (314) 747-4392; E-mail: [email protected] doi:10.1111/ijlh.12244
Received 24 September 2013; accepted for publication 26 February 2014 Keywords Flow cytometry, quality assurance, hematopathology, lymphoma
S U M M A RY Introduction: The utility of flow cytometry (FC) in diagnosis and staging of hematologic malignancy is controversial. Often, multiple specimens from the same patient are processed concurrently for FC analyses, alongside tissue for histomorphologic diagnosis. Methods: To assess the diagnostic utility of multiple, concurrent FC analyses, a 10-year retrospective review of cases with ≥2 concurrent specimens (from the same patient) submitted for FC was conducted. Light microscopic (LM) diagnoses were compared to FC findings, and the contribution of FC results to final diagnoses was examined. Results: Of 4058 specimens (predominantly lymph nodes, bone marrows, and oropharyngeal tissues) submitted for FC analyses, 129 (3.2%) represented cases with multiple (average: 2.19) concurrent FC analyses. All were accompanied by tissues and/or aspirates for LM examination. In 115 (89.1%) cases, multiple FC analyses were performed prior to morphologic examination. In 87.0% of those cases, ≥1 FC result(s) aligned with LM findings. In 15 (13.0%) cases where FC results differed from morphologic diagnoses, 86.7% (13/15) failed to detect an abnormal cell population by FC in the presence of a hematologic malignancy by LM. In one case (0.9%), FC detected a lymphoma, without morphologic evidence by LM. Conclusions: Overall, multipart FC failed to demonstrate a significant contribution in initial diagnoses of hematologic malignancies compared with analysis of a single specimen.
INTRODUCTION Staging is a critical component of the diagnosis of hematologic malignancy. The World Health Organization (WHO) categorizes and stages hematologic malignancies based on specific tissue morphologies,
90
immunophenotypes, genetic alterations, and clinical features [1]. Although morphology may be influenced by tumor size and viability [2], tissue size and sampling [3], and/or underlying disease processes (e.g., fibrosis, sclerosis, inflammation) confounding diagnoses [4], cellular or tissue morphology is considered the
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
gold standard for diagnosis [5]. However, certain hematologic malignancies have specific aberrant immunophenotypes, which are helpful in accurately recognizing an entity [1, 6]. Consequently, diagnostic evaluation of specimens suspicious for involvement by a hematologic malignancy increasingly includes flow cytometric, immunohistochemical, and/or molecular genetic analyses, as important adjuncts to morphologic assessment. In general, requests for both light microscopic (LM) morphologic examination and ancillary testing are made concurrently, and flow cytometric, immunohistochemical, and/or molecular genetic results are typically integrated into the final reports alongside the LM diagnoses. Several studies have examined the utility of ancillary techniques in initial diagnosis and disease monitoring in hematologic malignancies. Flow cytometry (FC) allows detection of multiple fluorescent markers simultaneously [6], which may facilitate narrowing of a broad differential over an abbreviated time course. Distinct cell populations are sorted by size and cytoplasmic granularity, and degenerating or nonviable cells can be excluded from the analyses [6]. The resultant data are quantitative and less prone to subjective bias compared with histomorphologic assessment [1] and have greater sensitivities and improved turnaround times as compared to immunohistochemical staining [7]. In addition, FC may be applied to cellular aspirates and body fluids, potentially obviating the need for a more invasive biopsy procedure [6]. In certain circumstances, however, FC does not render useful diagnostic information. In sclerotic, hypocellular, or fibrotic specimens, aspiration techniques often yield inadequate cellularity for assessment [6]. The relationship of cytologic features within their morphologic architecture is lost. Certain hematologic malignancies (i.e., Hodgkin Disease and T-cell lymphomas without genetic aberrations) typically evade flow cytometric detection due to the paucicellularity of neoplastic cells or the preservation of normal surface antigen expression [6]. In most institutions, results of FC analyses, as well as other ancillary studies, are interpreted within the context of the tissue or cellular morphologic phenotypes. Numerous studies have examined the role of FC in the initial diagnosis and monitoring of disease recurrence over a wide range of hematologic malignancies [1, 6, 8–10], including follicular lymphomas [11, 12], diffuse large © 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
91
B-cell lymphomas [12, 13], Mantle cell lymphomas [10], multiple myelomas, clonal plasma cell disorders [14], and other B-cell non-Hodgkin lymphomas [5] with mixed results. Only one study, to our knowledge, examined the utility of performing multiple ancillary studies concurrently on specimens from the same patient. Wang et al. [15] evaluated the need for concomitant histomorphologic, as well as flow cytometric, cytogenetic, and molecular, examination of bilateral bone marrow biopsies submitted concurrently as part of routine diagnostic evaluation for non-Hodgkin lymphoma (NHL), carcinoma, Hodgkin lymphoma (HL), sarcoma, myeloma, acute and chronic leukemia, myelodysplasia, and myeloproliferative disorders. Their results confirmed that LM examination of bilateral bone marrow biopsies increased the sensitivity of detecting neoplastic processes in patients with certain malignancies (NHL, HL, carcinoma, and sarcoma); however, none of the ancillary techniques, including FC, provided additional information that justified bilateral analyses [15]. To our knowledge, no study has evaluated the utility of multipart, concurrent FC in other tissue types. Here, we assess the diagnostic and clinical utility of multipart (≥2 specimens submitted concurrently) FC performed on a variety of specimen types (lymph node, bone marrow, oropharyngeal tissue, gastrointestinal tissue, respiratory tissue, soft tissue, bone, nasopharyngeal tissue, spleen, liver, kidney, thymus, parotid gland, and prostate) over a 10-year period. Although not explicitly indicated, multiple specimens were submitted for concurrent flow cytometric analysis to facilitate diagnosis or as part of presumed disease staging.
M AT E R I A L S A N D M E T H O D S To examine the contribution of multi-part (≥2 specimens submitted concurrently) FC analyses, a 10-year retrospective review of specimens submitted to the Lauren V. Ackerman Department of Surgical Pathology Laboratory for flow cytometric analysis between July 1, 2002, and July 1, 2012, was performed. Final reports from cases with multiple (≥2) specimens submitted for concurrent FC were further reviewed. Diagnoses rendered from LM examination of histo- or cytomorphology of tissues or aspirate smears were compared to findings from each specimen’s flow cytometric analysis. Concordance between LM diagnoses and flow cytomet-
92
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
ric findings, as well as across independent flow cytometric analyses, was determined. Discordant diagnoses were characterized by disease process. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of FC were calculated. This study was approved by the institutional review board (IRB) at Washington University School of Medicine, in accordance with the ethical standards established by the Helsinki Declaration of 1975.
(a)
R E S U LT S
(b)
From July 1, 2002, to July 1, 2012, 4058 specimens were submitted for flow cytometric analysis at our institution, of which 129 (3.2%) cases (from 128 patients) included requests for concurrent FC on multiple (≥2) specimens. In 14 cases (10.9%), FC was either not performed on any specimen submitted (n = 4) due to insufficient material or morphologic evidence of benignity, or was performed only on one of the specimens submitted (n = 10). These 14 cases were excluded from further analysis. Of the 115 (89.1%) cases on which multiple flow cytometric analyses were requested and performed, 98 cases (85.2%) were submitted for analysis of two specimens, 14 cases (12.2%) were submitted for analysis of three specimens, two cases (1.7%) were submitted for analysis of four specimens, and one case (0.9%) was submitted for analysis of five specimens (Figure 1a). In total, 251 flow cytometric analyses were performed on independent specimens. Tissue types submitted for analyses originated from a variety of organ systems (Figure 1b). Lymph node specimens were most frequently submitted (n = 120, 48%), with bone marrow (47; 19%) and oropharyngeal (tonsillar, base of tongue, valeculla) tissues (28; 11%) also well represented. Other tissue types submitted for analysis included gastrointestinal (including mucosa, as well as mesentery and omentum) specimens (n = 13) and respiratory tissues (n = 11), as well as soft tissues (n = 6), bone (n = 6), nasopharynx (n = 5), spleen (n = 4), kidney (n = 4), liver (n = 4), thymus (n = 1) prostate (n = 1), and parotid gland (n = 1; Figure 1b). No peripheral blood specimens were represented in our data set. The vast majority of specimens submitted for concurrent flow cytometric analyses (n = 82; 71%) included multiple samples from a solitary mass, multiple oropharyngeal tissue specimens (i.e., bilateral tonsils, base of tongue), or multiple lymph node
Figure 1. (a) Pie chart depicting the number of concurrent flow cytometric (FC) analyses performed per case. Of 115 cases on which multiple, concurrent FC analyses were performed, 98 cases (85.2%) were submitted for analysis of two specimens, 14 cases (12.2%) were submitted for analysis of three specimens, two cases (1.7%) were submitted for analysis of four specimens, and one case (0.9%) was submitted for analysis of five specimens. (b) Pie chart depicting the types of specimens submitted for multiple, concurrent FC analyses. Specimens from multiple organ systems were submitted, including lymph nodes (n = 120, 48%), bone marrow (n = 47; 19%), oropharyngeal (tonsillar, base of tongue, valeculla) tissues (n = 28; 11%), gastrointestinal (including mucosa, as well as mesentery and omentum) tissues (n = 13; 5%), respiratory tissues (n = 11; 4%), soft tissues (n = 6), bone (n = 6), nasopharynx (n = 5), spleen (n = 4), kidney (n = 4), liver (n = 4), thymus (n = 1) prostate (n = 1), and parotid gland (n = 1).
specimens, presumably to facilitate diagnosis and/or to assist in staging of potential malignancies. These specimens were derived from the same site (if extranodal) or from the same lymph node distribution. The remaining cases (n = 33; 29%) represented bilateral posterior iliac crest bone marrow biopsies or a mass lesion paired with an associated lymph node, presumably for use in tumor staging. © 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
Final diagnoses were reviewed for all cases on which multipart, concurrent FC was performed. In 100 of 115 (87.0%) cases, ≥1 FC result(s) agreed with LM findings. Of the 100 cases with concordant morphologic and flow cytometric results, 75 cases (75%) showed benign findings with no evidence of malignancy on LM examination of morphology and no monotypic B-cell population by FC. Seven (7%) of the concordant cases represented nonhematologic malignancies, including two cases of squamous cell carcinoma, one case of nasopharyngeal carcinoma, one prostatic adenocarcinoma, one seminoma, one epithelioid hemangioendothelioma, and one small round blue cell tumor. The remaining 18 concordant cases included 16 cases (16%) of non-Hodgkin B-cell lymphoma, including follicular lymphoma (n = 5; 5%), diffuse large B-cell lymphoma (n = 3; 3%), marginal zone lymphoma (n = 1; 1%), small lymphocytic lymphoma/chronic lymphocytic leukemia (n = 1, 1%), extracavitary primary effusion lymphoma (n = 1, 1%), and lymphoma not otherwise specified (n = 5; 5%), as well as two cases (2%) of acute myelogenous leukemia (Table 1). In 15 (13.0%) cases, flow cytometric results differed from morphologic diagnoses. The discordant cases included six cases of Hodgkin lymphoma (for which the aberrant B-cell population is not routinely identifiable by FC), two cases of diffuse large B-cell lymphoma, one case of follicular lymphoma, one case of B-lymphoblastic lymphoma, one case of post-transplant lymphoproliferative disorder, one natural killer (NK) cell neoplasm, one case of myelodysplastic syndrome (MDS), and one case of T-cell large granular lymphoma. In addition, one case that was negative for malignancy by light microscopy showed a small monotypic B-cell population by FC, the significance of which is unknown. In 13 of these 15 discordant cases (86.7%), the final diagnoses were rendered based on LM examination of morphologic features, with negative flow cytometric findings. In just one case, where T-cell large granular lymphoma was diagnosed, did flow cytometric findings influence the diagnosis without LM evidence of disease. Based on these data, the sensitivity, specificity, PPV, and NPV of flow cytometric analyses were calculated. The sensitivity in our study was low (47.4%), suggesting that FC is not an appropriate screening test to rule out hematologic malignancy prior to © 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
93
Table 1. Cases with multipart, concurrent flow cytometry M+/FC+ Negative for malignancy Benign, no clonal B-cell population Carcinoma Squamous cell carcinoma Nasopharyngeal carcinoma Prostatic adenocarcinoma Sarcoma Epithelioid hemangioendothelioma Germ cell tumor Seminoma Other nonhematologic malignancy Small round blue cell tumor B-cell non-Hodgkin lymphoma Diffuse large B-cell lymphoma Follicular lymphoma Marginal zone lymphoma Small lymphocytic lymphoma/chronic lymphocytic leukemia B-lymphoblastic lymphoma Post-transplant lymphoproliferative disorder Extracavitary primary effusion lymphoma Not otherwise specified Hodgkin lymphoma T-cell neoplasm T-cell large granular lymphoma Natural killer (NK) cell neoplasm Acute myelogenous leukemia Myelodysplastic syndrome (MDS)
M /FC
M+/FC
75
M /FC+
1
2 1 1
1
1
1
3
2
5 1
1
1
1 1
1
5 6 1 1 2 1
morphologic assessment via light microscopy. The specificity of FC was high (97.4%), suggesting that FC is better utilized to support a morphologic impression,
94
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
when confirming the presence of a hematologic malignancy. The PPV was high (90%), suggesting that a positive flow cytometric result was highly correlated with the presence of disease. The NPV was only 78.9%, suggesting that a negative flow cytometric result may not accurately reflect the absence of disease (Figure 2a). Interestingly, when the results of flow cytometric analysis were negative in at least one specimen, the results were negative in all concurrent specimens of a case, regardless of whether the morphologic findings were ultimately deemed negative or positive (Figure 2b). In cases with at least one positive (a)
(b)
Figure 2. (a) Two by two table comparing flow cytometric (FC) testing to morphologic diagnoses. In our study, the sensitivity of FC as an ancillary test was low (47.7%), and the specificity was high (97.4%). The positive predictive value was high (90%), while the negative predictive value was 78.9%. (b) Two by two table illustrating percent concordance (%) across concurrent FC analyses in relation to morphologic diagnoses. Cases with at least one negative result by FC showed negative, concordant FC results across all independent specimens, regardless of the corresponding morphologic diagnoses. Cases with at least one positive FC result showed 67% concordance across independent FC analyses when the morphologic diagnosis was positive, and only 50% concordance when the morphologic diagnosis was negative.
result by FC, 67% of cases showed positive, concordant results morphologically, while 50% of cases were negative by morphologic inspection (Figure 2b). To address whether performing multiple (>1), concurrent flow cytometric analyses significantly impacts diagnoses, the concordance across independent flow cytometric studies performed on the same patient was examined. Interestingly, when two specimens were submitted for concurrent flow cytometric analyses, the resultant outputs were concordant 95% of the time. When three specimens were submitted, the concordance across flow cytometric results was 86%. In the two cases where four specimens were submitted, as well as in the single case where five specimens were submitted for concurrent flow cytometric analyses, the results were 100% concordant (Table 2), suggesting that additional concurrent flow cytometric analyses do not contribute significantly to the final diagnosis. Last, the concordance across concurrent flow cytometric analyses was examined by specific disease process. In all three cases of DLBCL, as well as in each single case of marginal zone lymphoma and extracavitary primary effusion lymphoma, the flow cytometric results were concordant and positive in all specimens analyzed (Table 3). Of five cases of follicular lymphoma, only 2 (40%) showed concordant flow cytometric analyses. Only one of the two cases (50%) of acute myelogenous leukemia showed concordant, positive flow cytometric results. And, in four of five cases (80%) of B-cell NHL were all flow cytometric results concordant (Table 3). In the single case of small lymphocytic leukemia/chronic lymphocytic lymphoma (SLL/CLL), where two independent specimens were submitted for concurrent morphologic and flow cytometric analyses, only one specimen showed concordant results across the two platforms. The second specimen was positive by morphologic assessment, but
Table 2. Concordance across flow cytometric (FC) analyses by specimen number. Number of FC specimens
Concordance across FC, %
2 3 4 5
95 86 100 100
(98 cases) (14 cases) (2 cases) (1 case)
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
failed to yield adequate cellularity for FC analysis (Table 3). In the single case where morphologic inspection showed reactive lymphoid hyperplasia, but FC detected a small, monotypic B-cell population, the flow cytometric results were concordant across all three specimens (Table 3). Finally, in the single case where T-cell large granular lymphoma was diagnosed by FC, but not seen morphologically, the diagnosis was rendered on one specimen only. The FC from the second, concurrent specimen failed to yield adequate cellularity for analysis (Table 3).
DISCUSSION Flow cytometric is routinely used in the diagnosis and immunophenotyping of hematologic malignancies [6]. Despite conflicting evidence to support its efficacy as an ancillary tool in initial diagnosis and staging [11, 16–20], FC is frequently requested concurrent with LM examination of tissue or cellular morphology. It is well established that increasing the number of specimens submitted for morphologic assessment increases sensitivity of detection of malignant diseases [21–24].
Table 3. Concordance across flow cytometric (FC) analyses by disease state. M+/FC+ B-NHL
DLBCL (%) Follicular lymphoma (%) Marginal zone (%) SLL/CLL (%) Extracavitary Primary Effusion Lymphoma (%) NOS (%) AML (%)
3 (100)
M /FC+ Reactive hyperplasia with small monotypic B-cell population (%) T-cell large granular (%)
1 (100)
1 (0)
5 (40)
1 (100) 1 (0) 1 (100)
5 (80) 2 (50)
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
95
This concept underlies the routine practice of bilateral bone marrow examinations in the workup of hematologic malignancy, such as NHL. However, it is not known whether the same approach applies to ancillary studies, like FC. One study, to our knowledge, examined the utility of multiple flow cytometric analyses performed on bilateral bone marrow biopsies from patients with a spectrum of hematologic malignancies and found no role for multiple flow cytometric analyses [15]. At our institution, a variety of tissue types, including lymph node, bone marrow, oropharynx, gastrointestinal tract, lung, nasopharynx, and soft tissue, are submitted for flow cytometric analysis. Often, multiple specimens from the same patient are submitted for independent FC processing. Here, we examined the utility of multipart FC in the initial diagnostic workup of patients with concern for hematologic malignancy. Our results demonstrate that in only one case (0.87%) did multiple FC analyses provide diagnostically relevant information, which was not seen on morphologic examination. In the vast majority (87.0%) of cases, results from ≥1 flow cytometric analysis were identical to those obtained from morphologic examination. In one case (0.87%) a small, lambda surface immunoglobulin light chain-restricted B-cell population was detected by FC, when reactive lymphoid hyperplasia was seen morphologically. In the remaining cases where flow cytometric and morphologic diagnoses diverged, final diagnoses of hematologic malignancy were rendered by morphology alone, as FC failed to demonstrate abnormal cell populations in any case. A limitation of this retrospective review is the difficulty in determining the degree to which concurrent flow cytometric results influenced morphologic diagnoses. However, it is likely that a benign morphologic impression was strengthened by negative FC results, and vice versa. In many cases, it is also unclear whether multiple flow cytometric specimens were submitted to facilitate diagnosis or to provide information for cancer staging. Seventy-one percent of flow cytometric specimens included multiple samplings of solitary masses, oropharyngeal tissues, and lymph nodes that were derived from the same or an adjacent site. It should be noted that in 13 of the 15 cases (86.7%) where flow cytometric and morphologic diagnoses were discordant, immunophenotypic characterization was pursued using immunohistochemical and in situ hybridization techniques. In one case of
96
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
Hodgkin lymphoma, these additional studies could not be performed, as only an aspirate smear was available for examination. In the case of the natural killer cell neoplasm, no additional stains were performed; however, the morphologic features seen by light microscopy were compared to the patient’s previous specimen, on which five immunohistochemical stains had been performed. On average, 8.77 (range: 4–17) immunohistochemical and/or in situ hybridization stains were applied to each case to facilitate diagnosis. Immunohistochemistry and in situ hybridization studies are frequently employed in the workup of hematologic malignancies. Typically, however, these techniques are applied after morphologic review of the main specimen. Results are then interpreted after development of an LM differential diagnosis, as well as within the context of the tissue architecture. Although less amenable to multiple antigen labeling and less quantitative than FC, immunohistochemistry is better suited for characterizing malignancies with fibrous or sclerotic stroma [1]. It is also preferred over FC in cases suspicious for involvement by Hodgkin Lymphoma [1]. A final consideration in assessing the utility of FC relates to the cost of testing. A standard (2–8 antibody panel) FC workup bears substantial cost to the patient. At our institution, the combined technical and professional billing rates for a single standard FC panel are $436. The cost increases to $491 for a 9–15 antibody panel and to $586 for testing of >15 antibodies. As cost conscious and quality improvement measures are rapidly being integrated into healthcare, it is prudent to consider the financial aspects of the tests performed in diagnostic workups, particularly when multiples of these tests are requested. Clinicians routinely submit
REFERENCES 1. Stetler-Stevenson M. Flow cytometry in lymphoma diagnosis and prognosis: useful? Best Pract Res Clin Haematol 2003;16:583–97. 2. Bain BJ. Bone marrow trephine biopsy. J Clin Pathol 2001;54:737–42. 3. Bishop PW, McNally K, Harris M. Audit of bone marrow trephines. J Clin Pathol 1992;45:1105–8. 4. Vega F, Medeiros LJ, Lang WH, Mansoor A, Bueso-Ramos C, Jones D. The stromal
tissue(s) for flow cytometric analyses (as well as for a number of other potential ancillary tests) to facilitate narrowing of a broad clinical differential diagnosis. It is the role of the pathologist, however, to apply these ancillary tests in the appropriate morphologic or pathologic context. A recent article by Karen Titus described the successful redevelopment of one hematopathology laboratory’s workflow so as to ensure morphologic assessment of tissue prior to ancillary testing: ‘if a pathologist looks at the morphology of the bone marrow, [the] differential diagnosis goes from huge to small, very, very fast. The problem comes when ancillary studies are done reflexively [25].’ Our data suggest that single-part FC is best utilized as an ancillary test to support a morphologic impression. There is no evidence to support multipart, concurrent flow cytometric analyses (specifically on more than one specimen per case) prior to LM review. When used selectively to confirm a morphologic diagnosis, FC has high specificity. Specimens submitted for FC should therefore be held until morphologic review to ensure appropriate applicability of the test.
AC K N OW L E D G E M E N T S The authors thank Charles Eby, MD, for his assistance in and critical review of our data analyses. This work was supported by the Washington University Department of Pathology.
DISCLOSURE The authors attest that neither individual has financial disclosures.
composition of malignant lymphoid aggregates in bone marrow: variations in architecture and phenotype in different B-cell tumours. Br J Haematol 2002;117: 569–76. 5. Stacchini A, Demurtas A, Godio L, Martini G, Antinoro V, Palestro G. Flow cytometry in the bone marrow staging of mature Bcell neoplasms. Cytometry B Clin Cytom 2003;54:10–8. 6. Dunphy CH. Applications of flow cytometry and immunohistochemistry to diagnos-
tic hematopathology. Arch Pathol Lab Med 2004;128:1004–22. 7. Tbakhi A, Edinger M, Myles J, Pohlman B, Tubbs RR. Flow cytometric immunophenotyping of non-Hodgkin’s lymphomas and related disorders. Cytometry 1996;25:113– 24. 8. Banerjee D. Flow cytometry related to hematopoietic malignancy. Transfus Sci 1995;16:315–20. 9. Merli M, Arcaini L, Boveri E, Rattotti S, Picone C, Passamonti F, Tenore A, Sozzani
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
10.
11.
12.
13.
14.
L, Lucioni M, Varettoni M, Rizzi S, Morello L, Ferretti V, Pascutto C, Paulli M, Lazzarino M. Assessment of bone marrow involvement in non-Hodgkin’s lymphomas: comparison between histology and flow cytometry. Eur J Haematol 2010;85:405–15. Perea G, Altes A, Bellido M, Aventin A, Bordes R, Ayats R, Remacha AF, Espinosa I, Briones J, Sierra J, Nomdedeu JF. Clinical utility of bone marrow flow cytometry in B-cell non-Hodgkin lymphomas (BNHL). Histopathology 2004;45:268–74. Iancu D, Hao S, Lin P, Anderson SK, Jorgensen JL, McLaughlin P, Medeiros LJ. Follicular lymphoma in staging bone marrow specimens: correlation of histologic findings with the results of flow cytometry immunophenotypic analysis. Arch Pathol Lab Med 2007;131:282–7. Palacio C, Acebedo G, Navarrete M, RuizMarcellan C, Sanchez C, Blanco A, Lopez A. Flow cytometry in the bone marrow evaluation of follicular and diffuse large B-cell lymphomas. Haematologica 2001;86:934–40. Talaulikar D, Dahlstrom JE. Staging bone marrow in diffuse large B-cell lymphoma: the role of ancillary investigations. Pathology 2009;41:214–22. Paiva B, Almeida J, Perez-Andres M, Mateo G, Lopez A, Rasillo A, Vidriales MB, Lopez-
15.
16.
17.
18.
19.
Berges MC, Miguel JF, Orfao A. Utility of flow cytometry immunophenotyping in multiple myeloma and other clonal plasma cell-related disorders. Cytometry B Clin Cytom 2010;78:239–52. Wang J, Weiss LM, Chang KL, Slovak ML, Gaal K, Forman SJ, Arber DA. Diagnostic utility of bilateral bone marrow examination: significance of morphologic and ancillary technique study in malignancy. Cancer 2002;94:1522–31. Duggan PR, Easton D, Luider J, Auer IA. Bone marrow staging of patients with nonHodgkin lymphoma by flow cytometry: correlation with morphology. Cancer 2000;88:894–9. Hanson CA, Kurtin PJ, Katzmann JA, Hoyer JD, Li CY, Hodnefield JM, Meyers CH, Habermann TM, Witzig TE. Immunophenotypic analysis of peripheral blood and bone marrow in the staging of B-cell malignant lymphoma. Blood 1999;94:3889–96. Naughton MJ, Hess JL, Zutter MM, Bartlett NL. Bone marrow staging in patients with non-Hodgkin’s lymphoma: is flow cytometry a useful test? Cancer 1998;82:1154–9. Sandhaus LM, Voelkerding KV, Dougherty J, Raska K Jr. Combined utility of gene rearrangement analysis and flow cytometry in the diagnosis of lymphoproliferative dis-
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
20.
21.
22.
23.
24.
25.
97
ease in the bone marrow. Hematol Pathol 1990;4:135–48. Wells DA, Hall MC, Shulman HM, Loken MR. Occult B cell malignancies can be detected by three-color flow cytometry in patients with cytopenias. Leukemia 1998;12:2015–23. Brunning RD, Bloomfield CD, McKenna RW, Peterson LA. Bilateral trephine bone marrow biopsies in lymphoma and other neoplastic diseases. Ann Intern Med 1975;82:365–6. Juneja SK, Wolf MM, Cooper IA. Value of bilateral bone marrow biopsy specimens in non-Hodgkin’s lymphoma. J Clin Pathol 1990;43:630–2. Luoni M, Declich P, De Paoli A, Fava S, Marinoni P, Montalbetti L, Sangalli G, Sciuccati P, Tocci A, Tosi A. Bone marrow biopsy for the staging of non-Hodgkin’s lymphoma: bilateral or unilateral trephine biopsy? Tumori 1995;81:410–3. Menon NC, Buchanan JG. Bilateral trephine bone marrow biopsies in Hodgkin’s and non-Hodgkin’s lymphoma. Pathology 1979;11:53–7. Titus K. Lab teams up to curb unneeded testing. In: College of American Pathologists Today. Sherrie L. Rice. (ed.).Northfield, IL: BPA Worldwide; 2012: 1–43.
Copyright of International Journal of Laboratory Hematology is the property of WileyBlackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
INTERNAT IONAL JOURNAL OF LABORATO RY HEMATO LOGY
A quality improvement assessment of multiple, concurrent flow cytometry analyses at a tertiary care center H. E. KARNES, J. L. FRATER
Department of Pathology & Immunology, Washington University in St. Louis School of Medicine, St. Louis, MO, USA Correspondence: John L. Frater, Department of Pathology & Immunology, Washington University in St. Louis School of Medicine, 660 S. Euclid Avenue, Campus Box 8118, St. Louis, MO 63110, USA. Tel.: (314) 362-1553; Fax: (314) 747-4392; E-mail: [email protected] doi:10.1111/ijlh.12244
Received 24 September 2013; accepted for publication 26 February 2014 Keywords Flow cytometry, quality assurance, hematopathology, lymphoma
S U M M A RY Introduction: The utility of flow cytometry (FC) in diagnosis and staging of hematologic malignancy is controversial. Often, multiple specimens from the same patient are processed concurrently for FC analyses, alongside tissue for histomorphologic diagnosis. Methods: To assess the diagnostic utility of multiple, concurrent FC analyses, a 10-year retrospective review of cases with ≥2 concurrent specimens (from the same patient) submitted for FC was conducted. Light microscopic (LM) diagnoses were compared to FC findings, and the contribution of FC results to final diagnoses was examined. Results: Of 4058 specimens (predominantly lymph nodes, bone marrows, and oropharyngeal tissues) submitted for FC analyses, 129 (3.2%) represented cases with multiple (average: 2.19) concurrent FC analyses. All were accompanied by tissues and/or aspirates for LM examination. In 115 (89.1%) cases, multiple FC analyses were performed prior to morphologic examination. In 87.0% of those cases, ≥1 FC result(s) aligned with LM findings. In 15 (13.0%) cases where FC results differed from morphologic diagnoses, 86.7% (13/15) failed to detect an abnormal cell population by FC in the presence of a hematologic malignancy by LM. In one case (0.9%), FC detected a lymphoma, without morphologic evidence by LM. Conclusions: Overall, multipart FC failed to demonstrate a significant contribution in initial diagnoses of hematologic malignancies compared with analysis of a single specimen.
INTRODUCTION Staging is a critical component of the diagnosis of hematologic malignancy. The World Health Organization (WHO) categorizes and stages hematologic malignancies based on specific tissue morphologies,
90
immunophenotypes, genetic alterations, and clinical features [1]. Although morphology may be influenced by tumor size and viability [2], tissue size and sampling [3], and/or underlying disease processes (e.g., fibrosis, sclerosis, inflammation) confounding diagnoses [4], cellular or tissue morphology is considered the
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
gold standard for diagnosis [5]. However, certain hematologic malignancies have specific aberrant immunophenotypes, which are helpful in accurately recognizing an entity [1, 6]. Consequently, diagnostic evaluation of specimens suspicious for involvement by a hematologic malignancy increasingly includes flow cytometric, immunohistochemical, and/or molecular genetic analyses, as important adjuncts to morphologic assessment. In general, requests for both light microscopic (LM) morphologic examination and ancillary testing are made concurrently, and flow cytometric, immunohistochemical, and/or molecular genetic results are typically integrated into the final reports alongside the LM diagnoses. Several studies have examined the utility of ancillary techniques in initial diagnosis and disease monitoring in hematologic malignancies. Flow cytometry (FC) allows detection of multiple fluorescent markers simultaneously [6], which may facilitate narrowing of a broad differential over an abbreviated time course. Distinct cell populations are sorted by size and cytoplasmic granularity, and degenerating or nonviable cells can be excluded from the analyses [6]. The resultant data are quantitative and less prone to subjective bias compared with histomorphologic assessment [1] and have greater sensitivities and improved turnaround times as compared to immunohistochemical staining [7]. In addition, FC may be applied to cellular aspirates and body fluids, potentially obviating the need for a more invasive biopsy procedure [6]. In certain circumstances, however, FC does not render useful diagnostic information. In sclerotic, hypocellular, or fibrotic specimens, aspiration techniques often yield inadequate cellularity for assessment [6]. The relationship of cytologic features within their morphologic architecture is lost. Certain hematologic malignancies (i.e., Hodgkin Disease and T-cell lymphomas without genetic aberrations) typically evade flow cytometric detection due to the paucicellularity of neoplastic cells or the preservation of normal surface antigen expression [6]. In most institutions, results of FC analyses, as well as other ancillary studies, are interpreted within the context of the tissue or cellular morphologic phenotypes. Numerous studies have examined the role of FC in the initial diagnosis and monitoring of disease recurrence over a wide range of hematologic malignancies [1, 6, 8–10], including follicular lymphomas [11, 12], diffuse large © 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
91
B-cell lymphomas [12, 13], Mantle cell lymphomas [10], multiple myelomas, clonal plasma cell disorders [14], and other B-cell non-Hodgkin lymphomas [5] with mixed results. Only one study, to our knowledge, examined the utility of performing multiple ancillary studies concurrently on specimens from the same patient. Wang et al. [15] evaluated the need for concomitant histomorphologic, as well as flow cytometric, cytogenetic, and molecular, examination of bilateral bone marrow biopsies submitted concurrently as part of routine diagnostic evaluation for non-Hodgkin lymphoma (NHL), carcinoma, Hodgkin lymphoma (HL), sarcoma, myeloma, acute and chronic leukemia, myelodysplasia, and myeloproliferative disorders. Their results confirmed that LM examination of bilateral bone marrow biopsies increased the sensitivity of detecting neoplastic processes in patients with certain malignancies (NHL, HL, carcinoma, and sarcoma); however, none of the ancillary techniques, including FC, provided additional information that justified bilateral analyses [15]. To our knowledge, no study has evaluated the utility of multipart, concurrent FC in other tissue types. Here, we assess the diagnostic and clinical utility of multipart (≥2 specimens submitted concurrently) FC performed on a variety of specimen types (lymph node, bone marrow, oropharyngeal tissue, gastrointestinal tissue, respiratory tissue, soft tissue, bone, nasopharyngeal tissue, spleen, liver, kidney, thymus, parotid gland, and prostate) over a 10-year period. Although not explicitly indicated, multiple specimens were submitted for concurrent flow cytometric analysis to facilitate diagnosis or as part of presumed disease staging.
M AT E R I A L S A N D M E T H O D S To examine the contribution of multi-part (≥2 specimens submitted concurrently) FC analyses, a 10-year retrospective review of specimens submitted to the Lauren V. Ackerman Department of Surgical Pathology Laboratory for flow cytometric analysis between July 1, 2002, and July 1, 2012, was performed. Final reports from cases with multiple (≥2) specimens submitted for concurrent FC were further reviewed. Diagnoses rendered from LM examination of histo- or cytomorphology of tissues or aspirate smears were compared to findings from each specimen’s flow cytometric analysis. Concordance between LM diagnoses and flow cytomet-
92
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
ric findings, as well as across independent flow cytometric analyses, was determined. Discordant diagnoses were characterized by disease process. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of FC were calculated. This study was approved by the institutional review board (IRB) at Washington University School of Medicine, in accordance with the ethical standards established by the Helsinki Declaration of 1975.
(a)
R E S U LT S
(b)
From July 1, 2002, to July 1, 2012, 4058 specimens were submitted for flow cytometric analysis at our institution, of which 129 (3.2%) cases (from 128 patients) included requests for concurrent FC on multiple (≥2) specimens. In 14 cases (10.9%), FC was either not performed on any specimen submitted (n = 4) due to insufficient material or morphologic evidence of benignity, or was performed only on one of the specimens submitted (n = 10). These 14 cases were excluded from further analysis. Of the 115 (89.1%) cases on which multiple flow cytometric analyses were requested and performed, 98 cases (85.2%) were submitted for analysis of two specimens, 14 cases (12.2%) were submitted for analysis of three specimens, two cases (1.7%) were submitted for analysis of four specimens, and one case (0.9%) was submitted for analysis of five specimens (Figure 1a). In total, 251 flow cytometric analyses were performed on independent specimens. Tissue types submitted for analyses originated from a variety of organ systems (Figure 1b). Lymph node specimens were most frequently submitted (n = 120, 48%), with bone marrow (47; 19%) and oropharyngeal (tonsillar, base of tongue, valeculla) tissues (28; 11%) also well represented. Other tissue types submitted for analysis included gastrointestinal (including mucosa, as well as mesentery and omentum) specimens (n = 13) and respiratory tissues (n = 11), as well as soft tissues (n = 6), bone (n = 6), nasopharynx (n = 5), spleen (n = 4), kidney (n = 4), liver (n = 4), thymus (n = 1) prostate (n = 1), and parotid gland (n = 1; Figure 1b). No peripheral blood specimens were represented in our data set. The vast majority of specimens submitted for concurrent flow cytometric analyses (n = 82; 71%) included multiple samples from a solitary mass, multiple oropharyngeal tissue specimens (i.e., bilateral tonsils, base of tongue), or multiple lymph node
Figure 1. (a) Pie chart depicting the number of concurrent flow cytometric (FC) analyses performed per case. Of 115 cases on which multiple, concurrent FC analyses were performed, 98 cases (85.2%) were submitted for analysis of two specimens, 14 cases (12.2%) were submitted for analysis of three specimens, two cases (1.7%) were submitted for analysis of four specimens, and one case (0.9%) was submitted for analysis of five specimens. (b) Pie chart depicting the types of specimens submitted for multiple, concurrent FC analyses. Specimens from multiple organ systems were submitted, including lymph nodes (n = 120, 48%), bone marrow (n = 47; 19%), oropharyngeal (tonsillar, base of tongue, valeculla) tissues (n = 28; 11%), gastrointestinal (including mucosa, as well as mesentery and omentum) tissues (n = 13; 5%), respiratory tissues (n = 11; 4%), soft tissues (n = 6), bone (n = 6), nasopharynx (n = 5), spleen (n = 4), kidney (n = 4), liver (n = 4), thymus (n = 1) prostate (n = 1), and parotid gland (n = 1).
specimens, presumably to facilitate diagnosis and/or to assist in staging of potential malignancies. These specimens were derived from the same site (if extranodal) or from the same lymph node distribution. The remaining cases (n = 33; 29%) represented bilateral posterior iliac crest bone marrow biopsies or a mass lesion paired with an associated lymph node, presumably for use in tumor staging. © 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
Final diagnoses were reviewed for all cases on which multipart, concurrent FC was performed. In 100 of 115 (87.0%) cases, ≥1 FC result(s) agreed with LM findings. Of the 100 cases with concordant morphologic and flow cytometric results, 75 cases (75%) showed benign findings with no evidence of malignancy on LM examination of morphology and no monotypic B-cell population by FC. Seven (7%) of the concordant cases represented nonhematologic malignancies, including two cases of squamous cell carcinoma, one case of nasopharyngeal carcinoma, one prostatic adenocarcinoma, one seminoma, one epithelioid hemangioendothelioma, and one small round blue cell tumor. The remaining 18 concordant cases included 16 cases (16%) of non-Hodgkin B-cell lymphoma, including follicular lymphoma (n = 5; 5%), diffuse large B-cell lymphoma (n = 3; 3%), marginal zone lymphoma (n = 1; 1%), small lymphocytic lymphoma/chronic lymphocytic leukemia (n = 1, 1%), extracavitary primary effusion lymphoma (n = 1, 1%), and lymphoma not otherwise specified (n = 5; 5%), as well as two cases (2%) of acute myelogenous leukemia (Table 1). In 15 (13.0%) cases, flow cytometric results differed from morphologic diagnoses. The discordant cases included six cases of Hodgkin lymphoma (for which the aberrant B-cell population is not routinely identifiable by FC), two cases of diffuse large B-cell lymphoma, one case of follicular lymphoma, one case of B-lymphoblastic lymphoma, one case of post-transplant lymphoproliferative disorder, one natural killer (NK) cell neoplasm, one case of myelodysplastic syndrome (MDS), and one case of T-cell large granular lymphoma. In addition, one case that was negative for malignancy by light microscopy showed a small monotypic B-cell population by FC, the significance of which is unknown. In 13 of these 15 discordant cases (86.7%), the final diagnoses were rendered based on LM examination of morphologic features, with negative flow cytometric findings. In just one case, where T-cell large granular lymphoma was diagnosed, did flow cytometric findings influence the diagnosis without LM evidence of disease. Based on these data, the sensitivity, specificity, PPV, and NPV of flow cytometric analyses were calculated. The sensitivity in our study was low (47.4%), suggesting that FC is not an appropriate screening test to rule out hematologic malignancy prior to © 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
93
Table 1. Cases with multipart, concurrent flow cytometry M+/FC+ Negative for malignancy Benign, no clonal B-cell population Carcinoma Squamous cell carcinoma Nasopharyngeal carcinoma Prostatic adenocarcinoma Sarcoma Epithelioid hemangioendothelioma Germ cell tumor Seminoma Other nonhematologic malignancy Small round blue cell tumor B-cell non-Hodgkin lymphoma Diffuse large B-cell lymphoma Follicular lymphoma Marginal zone lymphoma Small lymphocytic lymphoma/chronic lymphocytic leukemia B-lymphoblastic lymphoma Post-transplant lymphoproliferative disorder Extracavitary primary effusion lymphoma Not otherwise specified Hodgkin lymphoma T-cell neoplasm T-cell large granular lymphoma Natural killer (NK) cell neoplasm Acute myelogenous leukemia Myelodysplastic syndrome (MDS)
M /FC
M+/FC
75
M /FC+
1
2 1 1
1
1
1
3
2
5 1
1
1
1 1
1
5 6 1 1 2 1
morphologic assessment via light microscopy. The specificity of FC was high (97.4%), suggesting that FC is better utilized to support a morphologic impression,
94
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
when confirming the presence of a hematologic malignancy. The PPV was high (90%), suggesting that a positive flow cytometric result was highly correlated with the presence of disease. The NPV was only 78.9%, suggesting that a negative flow cytometric result may not accurately reflect the absence of disease (Figure 2a). Interestingly, when the results of flow cytometric analysis were negative in at least one specimen, the results were negative in all concurrent specimens of a case, regardless of whether the morphologic findings were ultimately deemed negative or positive (Figure 2b). In cases with at least one positive (a)
(b)
Figure 2. (a) Two by two table comparing flow cytometric (FC) testing to morphologic diagnoses. In our study, the sensitivity of FC as an ancillary test was low (47.7%), and the specificity was high (97.4%). The positive predictive value was high (90%), while the negative predictive value was 78.9%. (b) Two by two table illustrating percent concordance (%) across concurrent FC analyses in relation to morphologic diagnoses. Cases with at least one negative result by FC showed negative, concordant FC results across all independent specimens, regardless of the corresponding morphologic diagnoses. Cases with at least one positive FC result showed 67% concordance across independent FC analyses when the morphologic diagnosis was positive, and only 50% concordance when the morphologic diagnosis was negative.
result by FC, 67% of cases showed positive, concordant results morphologically, while 50% of cases were negative by morphologic inspection (Figure 2b). To address whether performing multiple (>1), concurrent flow cytometric analyses significantly impacts diagnoses, the concordance across independent flow cytometric studies performed on the same patient was examined. Interestingly, when two specimens were submitted for concurrent flow cytometric analyses, the resultant outputs were concordant 95% of the time. When three specimens were submitted, the concordance across flow cytometric results was 86%. In the two cases where four specimens were submitted, as well as in the single case where five specimens were submitted for concurrent flow cytometric analyses, the results were 100% concordant (Table 2), suggesting that additional concurrent flow cytometric analyses do not contribute significantly to the final diagnosis. Last, the concordance across concurrent flow cytometric analyses was examined by specific disease process. In all three cases of DLBCL, as well as in each single case of marginal zone lymphoma and extracavitary primary effusion lymphoma, the flow cytometric results were concordant and positive in all specimens analyzed (Table 3). Of five cases of follicular lymphoma, only 2 (40%) showed concordant flow cytometric analyses. Only one of the two cases (50%) of acute myelogenous leukemia showed concordant, positive flow cytometric results. And, in four of five cases (80%) of B-cell NHL were all flow cytometric results concordant (Table 3). In the single case of small lymphocytic leukemia/chronic lymphocytic lymphoma (SLL/CLL), where two independent specimens were submitted for concurrent morphologic and flow cytometric analyses, only one specimen showed concordant results across the two platforms. The second specimen was positive by morphologic assessment, but
Table 2. Concordance across flow cytometric (FC) analyses by specimen number. Number of FC specimens
Concordance across FC, %
2 3 4 5
95 86 100 100
(98 cases) (14 cases) (2 cases) (1 case)
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
failed to yield adequate cellularity for FC analysis (Table 3). In the single case where morphologic inspection showed reactive lymphoid hyperplasia, but FC detected a small, monotypic B-cell population, the flow cytometric results were concordant across all three specimens (Table 3). Finally, in the single case where T-cell large granular lymphoma was diagnosed by FC, but not seen morphologically, the diagnosis was rendered on one specimen only. The FC from the second, concurrent specimen failed to yield adequate cellularity for analysis (Table 3).
DISCUSSION Flow cytometric is routinely used in the diagnosis and immunophenotyping of hematologic malignancies [6]. Despite conflicting evidence to support its efficacy as an ancillary tool in initial diagnosis and staging [11, 16–20], FC is frequently requested concurrent with LM examination of tissue or cellular morphology. It is well established that increasing the number of specimens submitted for morphologic assessment increases sensitivity of detection of malignant diseases [21–24].
Table 3. Concordance across flow cytometric (FC) analyses by disease state. M+/FC+ B-NHL
DLBCL (%) Follicular lymphoma (%) Marginal zone (%) SLL/CLL (%) Extracavitary Primary Effusion Lymphoma (%) NOS (%) AML (%)
3 (100)
M /FC+ Reactive hyperplasia with small monotypic B-cell population (%) T-cell large granular (%)
1 (100)
1 (0)
5 (40)
1 (100) 1 (0) 1 (100)
5 (80) 2 (50)
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
95
This concept underlies the routine practice of bilateral bone marrow examinations in the workup of hematologic malignancy, such as NHL. However, it is not known whether the same approach applies to ancillary studies, like FC. One study, to our knowledge, examined the utility of multiple flow cytometric analyses performed on bilateral bone marrow biopsies from patients with a spectrum of hematologic malignancies and found no role for multiple flow cytometric analyses [15]. At our institution, a variety of tissue types, including lymph node, bone marrow, oropharynx, gastrointestinal tract, lung, nasopharynx, and soft tissue, are submitted for flow cytometric analysis. Often, multiple specimens from the same patient are submitted for independent FC processing. Here, we examined the utility of multipart FC in the initial diagnostic workup of patients with concern for hematologic malignancy. Our results demonstrate that in only one case (0.87%) did multiple FC analyses provide diagnostically relevant information, which was not seen on morphologic examination. In the vast majority (87.0%) of cases, results from ≥1 flow cytometric analysis were identical to those obtained from morphologic examination. In one case (0.87%) a small, lambda surface immunoglobulin light chain-restricted B-cell population was detected by FC, when reactive lymphoid hyperplasia was seen morphologically. In the remaining cases where flow cytometric and morphologic diagnoses diverged, final diagnoses of hematologic malignancy were rendered by morphology alone, as FC failed to demonstrate abnormal cell populations in any case. A limitation of this retrospective review is the difficulty in determining the degree to which concurrent flow cytometric results influenced morphologic diagnoses. However, it is likely that a benign morphologic impression was strengthened by negative FC results, and vice versa. In many cases, it is also unclear whether multiple flow cytometric specimens were submitted to facilitate diagnosis or to provide information for cancer staging. Seventy-one percent of flow cytometric specimens included multiple samplings of solitary masses, oropharyngeal tissues, and lymph nodes that were derived from the same or an adjacent site. It should be noted that in 13 of the 15 cases (86.7%) where flow cytometric and morphologic diagnoses were discordant, immunophenotypic characterization was pursued using immunohistochemical and in situ hybridization techniques. In one case of
96
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
Hodgkin lymphoma, these additional studies could not be performed, as only an aspirate smear was available for examination. In the case of the natural killer cell neoplasm, no additional stains were performed; however, the morphologic features seen by light microscopy were compared to the patient’s previous specimen, on which five immunohistochemical stains had been performed. On average, 8.77 (range: 4–17) immunohistochemical and/or in situ hybridization stains were applied to each case to facilitate diagnosis. Immunohistochemistry and in situ hybridization studies are frequently employed in the workup of hematologic malignancies. Typically, however, these techniques are applied after morphologic review of the main specimen. Results are then interpreted after development of an LM differential diagnosis, as well as within the context of the tissue architecture. Although less amenable to multiple antigen labeling and less quantitative than FC, immunohistochemistry is better suited for characterizing malignancies with fibrous or sclerotic stroma [1]. It is also preferred over FC in cases suspicious for involvement by Hodgkin Lymphoma [1]. A final consideration in assessing the utility of FC relates to the cost of testing. A standard (2–8 antibody panel) FC workup bears substantial cost to the patient. At our institution, the combined technical and professional billing rates for a single standard FC panel are $436. The cost increases to $491 for a 9–15 antibody panel and to $586 for testing of >15 antibodies. As cost conscious and quality improvement measures are rapidly being integrated into healthcare, it is prudent to consider the financial aspects of the tests performed in diagnostic workups, particularly when multiples of these tests are requested. Clinicians routinely submit
REFERENCES 1. Stetler-Stevenson M. Flow cytometry in lymphoma diagnosis and prognosis: useful? Best Pract Res Clin Haematol 2003;16:583–97. 2. Bain BJ. Bone marrow trephine biopsy. J Clin Pathol 2001;54:737–42. 3. Bishop PW, McNally K, Harris M. Audit of bone marrow trephines. J Clin Pathol 1992;45:1105–8. 4. Vega F, Medeiros LJ, Lang WH, Mansoor A, Bueso-Ramos C, Jones D. The stromal
tissue(s) for flow cytometric analyses (as well as for a number of other potential ancillary tests) to facilitate narrowing of a broad clinical differential diagnosis. It is the role of the pathologist, however, to apply these ancillary tests in the appropriate morphologic or pathologic context. A recent article by Karen Titus described the successful redevelopment of one hematopathology laboratory’s workflow so as to ensure morphologic assessment of tissue prior to ancillary testing: ‘if a pathologist looks at the morphology of the bone marrow, [the] differential diagnosis goes from huge to small, very, very fast. The problem comes when ancillary studies are done reflexively [25].’ Our data suggest that single-part FC is best utilized as an ancillary test to support a morphologic impression. There is no evidence to support multipart, concurrent flow cytometric analyses (specifically on more than one specimen per case) prior to LM review. When used selectively to confirm a morphologic diagnosis, FC has high specificity. Specimens submitted for FC should therefore be held until morphologic review to ensure appropriate applicability of the test.
AC K N OW L E D G E M E N T S The authors thank Charles Eby, MD, for his assistance in and critical review of our data analyses. This work was supported by the Washington University Department of Pathology.
DISCLOSURE The authors attest that neither individual has financial disclosures.
composition of malignant lymphoid aggregates in bone marrow: variations in architecture and phenotype in different B-cell tumours. Br J Haematol 2002;117: 569–76. 5. Stacchini A, Demurtas A, Godio L, Martini G, Antinoro V, Palestro G. Flow cytometry in the bone marrow staging of mature Bcell neoplasms. Cytometry B Clin Cytom 2003;54:10–8. 6. Dunphy CH. Applications of flow cytometry and immunohistochemistry to diagnos-
tic hematopathology. Arch Pathol Lab Med 2004;128:1004–22. 7. Tbakhi A, Edinger M, Myles J, Pohlman B, Tubbs RR. Flow cytometric immunophenotyping of non-Hodgkin’s lymphomas and related disorders. Cytometry 1996;25:113– 24. 8. Banerjee D. Flow cytometry related to hematopoietic malignancy. Transfus Sci 1995;16:315–20. 9. Merli M, Arcaini L, Boveri E, Rattotti S, Picone C, Passamonti F, Tenore A, Sozzani
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
H. E. KARNES AND J. L. FRATER | FLOW CYTOMETRY UTILIZATION
10.
11.
12.
13.
14.
L, Lucioni M, Varettoni M, Rizzi S, Morello L, Ferretti V, Pascutto C, Paulli M, Lazzarino M. Assessment of bone marrow involvement in non-Hodgkin’s lymphomas: comparison between histology and flow cytometry. Eur J Haematol 2010;85:405–15. Perea G, Altes A, Bellido M, Aventin A, Bordes R, Ayats R, Remacha AF, Espinosa I, Briones J, Sierra J, Nomdedeu JF. Clinical utility of bone marrow flow cytometry in B-cell non-Hodgkin lymphomas (BNHL). Histopathology 2004;45:268–74. Iancu D, Hao S, Lin P, Anderson SK, Jorgensen JL, McLaughlin P, Medeiros LJ. Follicular lymphoma in staging bone marrow specimens: correlation of histologic findings with the results of flow cytometry immunophenotypic analysis. Arch Pathol Lab Med 2007;131:282–7. Palacio C, Acebedo G, Navarrete M, RuizMarcellan C, Sanchez C, Blanco A, Lopez A. Flow cytometry in the bone marrow evaluation of follicular and diffuse large B-cell lymphomas. Haematologica 2001;86:934–40. Talaulikar D, Dahlstrom JE. Staging bone marrow in diffuse large B-cell lymphoma: the role of ancillary investigations. Pathology 2009;41:214–22. Paiva B, Almeida J, Perez-Andres M, Mateo G, Lopez A, Rasillo A, Vidriales MB, Lopez-
15.
16.
17.
18.
19.
Berges MC, Miguel JF, Orfao A. Utility of flow cytometry immunophenotyping in multiple myeloma and other clonal plasma cell-related disorders. Cytometry B Clin Cytom 2010;78:239–52. Wang J, Weiss LM, Chang KL, Slovak ML, Gaal K, Forman SJ, Arber DA. Diagnostic utility of bilateral bone marrow examination: significance of morphologic and ancillary technique study in malignancy. Cancer 2002;94:1522–31. Duggan PR, Easton D, Luider J, Auer IA. Bone marrow staging of patients with nonHodgkin lymphoma by flow cytometry: correlation with morphology. Cancer 2000;88:894–9. Hanson CA, Kurtin PJ, Katzmann JA, Hoyer JD, Li CY, Hodnefield JM, Meyers CH, Habermann TM, Witzig TE. Immunophenotypic analysis of peripheral blood and bone marrow in the staging of B-cell malignant lymphoma. Blood 1999;94:3889–96. Naughton MJ, Hess JL, Zutter MM, Bartlett NL. Bone marrow staging in patients with non-Hodgkin’s lymphoma: is flow cytometry a useful test? Cancer 1998;82:1154–9. Sandhaus LM, Voelkerding KV, Dougherty J, Raska K Jr. Combined utility of gene rearrangement analysis and flow cytometry in the diagnosis of lymphoproliferative dis-
© 2014 John Wiley & Sons Ltd, Int. Jnl. Lab. Hem. 2015, 37, 90–97
20.
21.
22.
23.
24.
25.
97
ease in the bone marrow. Hematol Pathol 1990;4:135–48. Wells DA, Hall MC, Shulman HM, Loken MR. Occult B cell malignancies can be detected by three-color flow cytometry in patients with cytopenias. Leukemia 1998;12:2015–23. Brunning RD, Bloomfield CD, McKenna RW, Peterson LA. Bilateral trephine bone marrow biopsies in lymphoma and other neoplastic diseases. Ann Intern Med 1975;82:365–6. Juneja SK, Wolf MM, Cooper IA. Value of bilateral bone marrow biopsy specimens in non-Hodgkin’s lymphoma. J Clin Pathol 1990;43:630–2. Luoni M, Declich P, De Paoli A, Fava S, Marinoni P, Montalbetti L, Sangalli G, Sciuccati P, Tocci A, Tosi A. Bone marrow biopsy for the staging of non-Hodgkin’s lymphoma: bilateral or unilateral trephine biopsy? Tumori 1995;81:410–3. Menon NC, Buchanan JG. Bilateral trephine bone marrow biopsies in Hodgkin’s and non-Hodgkin’s lymphoma. Pathology 1979;11:53–7. Titus K. Lab teams up to curb unneeded testing. In: College of American Pathologists Today. Sherrie L. Rice. (ed.).Northfield, IL: BPA Worldwide; 2012: 1–43.
Copyright of International Journal of Laboratory Hematology is the property of WileyBlackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
