Although CD30 testing is an established tool in the diagnostic workup of lymphomas, it is also emerging as a predictive biomarker that informs treatment. The current definition of CD30 positivity by immunohistochemistry is descriptive and based on reactivity in lymphomas that are defined by their universal strong expression of CD30, rather than any established threshold. Challenges include inconsistencies with preanalytic variables, tissue processing, pathologist readout, and with the pathologist and oncologist interpretation of reported results.
To develop and propose general best practice recommendations for reporting CD30 expression by immunohistochemistry in lymphoma biopsies to harmonize practices across institutions and facilitate assessment of its significance in clinical decision-making.
Following literature review and group discussion, the panel of 14 academic hematopathologists and 2 clinical/academic hematologists/oncologists divided into 3 working groups. Each working group was tasked with assessing CD30 testing by immunohistochemistry, CD30 expression readout, or CD30 expression interpretation.
Panel recommendations were reviewed and discussed. An online survey was conducted to confirm the consensus recommendations.
CD30 immunohistochemistry is required for all patients in whom classic Hodgkin lymphoma and any lymphoma within the spectrum of peripheral T-cell lymphoma are differential diagnostic considerations. The panel reinforced and summarized that immunohistochemistry is the preferred methodology and any degree of CD30 expression should be reported. For diagnostic purposes, the interpretation of CD30 expression should follow published guidelines. To inform therapeutic decisions, report estimated percent positive expression in tumor cells (or total cells where applicable) and record descriptively if nontumor cells are positive.
CD30 (also known as Ki-1 or TNFRSF8) is a transmembrane glycoprotein receptor and a member of the tumor necrosis factor receptor superfamily.1,2 These proteins are expressed primarily by immune cells to regulate complex and diverse functions, including inflammation, apoptosis, and cell proliferation, although the precise pathways are not fully understood.1–4 Activated B and T cells might express CD30, whereas expression in resting normal cells is minimal, making CD30 a therapeutic target for several lymphomas based on its expression.5
CD30 expression is essential to diagnose classic Hodgkin lymphoma (cHL) as well as non-Hodgkin lymphomas of T-cell lineage (specifically systemic anaplastic large cell lymphoma [ALCL], primary cutaneous ALCL, lymphomatoid papulosis, and breast implant–associated ALCL). Lymphoid cells carrying viral Epstein-Barr virus, human immunodeficiency virus (HIV), and human T-cell leukemia-lymphoma virus type 1 genomes also express high levels of CD30.6 CD30 is variably expressed in other T- and B-cell malignancies, such as diffuse large B-cell lymphoma, primary mediastinal (thymic) large B-cell lymphoma, follicular lymphoma, peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS), angioimmunoblastic T-cell lymphoma (AITL), adult T-cell leukemia/lymphoma, extranodal natural killer/T-cell lymphoma, mycosis fungoides, Sézary syndrome, and primary cutaneous γδ T-cell lymphoma.2,4,7–15 Among T-cell lymphomas that express CD30, the percentage of CD30 expression identified in all cells within a given case tends to be lower in AITL and mycosis fungoides (mycosis fungoides, <30%) but can be higher in PTCL-NOS (up to 100%).16–19 Table 1 describes a selection of CD30 immunohistochemistry (IHC) criteria to report CD30 expression in the literature.
CD30 expression is an established and important biomarker in the diagnosis of lymphomas. Recently it has also been used to inform treatment with specific agents, resulting in improved outcomes of patients with first relapsed or refractory systemic ALCL and cutaneous T-cell lymphoma (CTCL) and recently in front-line therapy for patients with PTCL. Patients with rare diseases, such as systemic mastocytosis, γδ T-cell lymphoma, and germ cell tumors, such as embryonal carcinoma, have also demonstrated responses to CD30-directed therapy.20–23
Guidelines for CD30 testing by IHC in lymphomas of T- and B-cell lineage are currently less well defined. Historically, the definition of CD30 positivity has relied on comparison with cHL and systemic ALCL that require diffuse and strong expression of CD30 in neoplastic cells. Because CD30 is often reported in a diagnosis-dependent fashion, once ALK-negative ALCL is excluded, CD30 may be reported as negative even if present in a significant number of cells. Recent developments in precision medicine have identified inconsistencies in testing and reporting of CD30 by IHC in lymphomas, including variability with preanalytic factors, tissue processing, pathologist reporting, and pathologist and oncologist interpretation. Essentially CD30, which has been used primarily as a diagnostic biomarker, is now also being used as a predictive biomarker, where defined thresholds are not universally agreed upon and are being applied without reference to the clinical and analytic sensitivity of the assay. Furthermore, the analytic sensitivity of the CD30 IHC assay, as applied in the diagnosis of lymphomas, is not known or controlled. In part, these best practice recommendations aim to support the separation of CD30 reporting for diagnostic purposes from the reporting needed for treatment selection.
Many studies have demonstrated clinical responses with anti-CD30–targeted therapy without consideration to CD30 expression in tumor cells, including cases of tumors without visually detectable CD30 by IHC.14,16,24–26 This apparent lack of correlation between CD30 expression and likelihood of response may be due to CD30-independent tumor killing mechanisms, including antibody-dependent cellular phagocytosis, immunogenic cell death, the bystander effect, and depletion of CD30-expressing T-regulatory cells.27 This paradigm has led researchers to design more studies to better characterize the identification and positivity level of CD30 expression testing to explore potential treatment options for patients with CD30-expressing lymphomas. Ultimately, from the clinician's perspective, the previously characterized disease-specific descriptions used for diagnostic purposes might be less relevant than the expression of CD30. Herein, we provide recommendations that are intended to help identify who to test for CD30 with the most feasible and reliable testing method, along with suggesting ways to set descriptive analytic sensitivity, pathologist readout, interpretation, and reporting of the results. Because of the transparency in the reporting, these recommendations will also have an impact on the oncologist's interpretation of the reported results.
PARTICIPANTS
A panel of hematopathologists and 2 clinical hematologists/oncologists from 14 academic institutions in North America was convened to conduct a review of current CD30 procedures and develop best practice recommendations based on available published data and professional experience. The outcome objective of the group was to publish CD30 reporting recommendations with the goal of offering practical advice for clinical hematologists/oncologists, hematopathologists, and pathologists involved in lymphoma care. Members of the panel were encouraged to provide additional details and dissenting opinions for discussion if evidence was unclear or not yet generated.
Recognizing the difficulty in employing a diagnostic biomarker as a predictive biomarker to inform treatment, the panel's recommendations reflect the best available evidence based on phase 2 or 3 clinical trials with majority expert agreement supported in practice. These recommendations propose a framework for testing and reporting CD30 that can be seamlessly applied in clinics, laboratories, and institutions. In describing an approach to identify those patients who would benefit from CD30 testing, either diagnostic or therapeutic, the recommendations aim at increasing the transparency of the results achieved by the CD30 IHC assay by disclosing the percentage of positive neoplastic cells and the presence or absence in the inflammatory background whenever possible. These recommendations make no claims on any specific threshold being clinically informative for targeted therapy. It is the responsibility of the treating physician or other health care provider to establish the optimal therapy for each patient. Essentially, these recommendations should provide more granular and specific information about CD30 expression in lymphoma, leaving the treating physician in charge of making patient care decisions in the context of published results of clinical trials, as well as the patient's individual situation and preferences.
CONSENSUS PROCESS
Following literature review and group discussion, the panel was divided into 3 working groups. Each working group was tasked with assessing a key focus area: CD30 testing, CD30 expression readout, or CD30 expression interpretation. Panel recommendations were reviewed, and key concepts were further discussed. Following the meeting, an online survey was conducted to confirm consensus recommendations made by the expert panel working groups (Supplemental Table 1, see supplemental digital content at https://meridian.allenpress.com/aplm in the January 2023 table of contents). All participants in the online survey participated in the expert panel.
RESULTS
Recommendations are summarized in Table 2 and discussed below.
Recommendation 1: Indications for CD30 Testing
For diagnostic purposes, CD30 testing by IHC is required for all patients with suspected cHL and PTCL. CD30 testing might also be useful for diagnosing subsets of large B-cell lymphomas, such as gray zone lymphoma or primary mediastinal large B-cell lymphoma. CD30 assessment is generally performed as part of a panel of markers included in the workup of extranodal T/NK-cell lymphomas, nasal type, and thus also can improve the overall diagnostic accuracy of this group of lymphomas.28
To best inform treatment decisions, CD30 testing is recommended for all patients with a diagnosis of T-cell lymphoma. In selected cases and if CD30-directed therapy is being considered, then CD30 testing is useful for relapsed/refractory diffuse large B-cell lymphoma or primary mediastinal large B-cell lymphoma. Some institutions and hematopathologists prefer to test for CD30 in the initial biopsy to provide options if needed for relapsed/refractory patients.
In patients with suspected AITL (and where available), dual staining for PAX-5/CD30 can be useful to exclude the presence of activated B immunoblasts in samples (and separate CD30+ T or B cells).6,29 The lymphoma cells in AITL are composed of small- to medium-sized neoplastic cells with varying degrees of CD30 expression intensity and can be difficult to discriminate from reactive, activated B cells.30,31 CD30 expression is typically detected in PTCLs with a higher proportion of large cell tumors versus small cell malignancies.
Patients with CTCL tend to exhibit heterogeneity with variable expression of immunohistochemical markers both within a biopsy and between biopsy sites from the same patient.32 Thus, for predictive biomarker purposes, at least 2 separate skin biopsies from different sites should be considered to best characterize CD30 expression.
If a clinical trial requests, or if clinical indication arises from therapy, then CD30 testing should be repeated after treatment in follow-up biopsies.
Recommendation 2: How to Test
IHC testing is the preferred methodology. Tonsil is recommended as a positive control for CD30, rather than cHL or ALCL samples, which might not provide information on the limit of detection. Although there are more sensitive methods available for detecting CD30 expression, such as multispectral imaging or computer-assisted evaluation, they are not widely available or used, and they lack standardization and benchmarks. Nevertheless, these more sensitive methods of CD30 testing have reported CD30 expression at levels that had previously been undetectable by IHC. Gene expression profiling has demonstrated expression of the CD30 gene in all samples tested with PTCL-NOS and AITL; however, this is not seen in IHC testing, where many cases appear negative by IHC.33 The gene expression profiling data with PTCL biopsies are similar to the multispectral imaging results in CTCL, where CD30 is found in almost all samples tested.24,33 In addition, a moderately strong correlation between CD30 by IHC and mRNA levels has been found.29,34 Clinical studies have demonstrated treatment responses using anti-CD30–targeted therapy even when CD30 was previously undetectable by IHC.26,35
The reliability of flow cytometry varies with the type of cancer; for example, for most cases of non-Hodgkin lymphoma, flow cytometry has high analytic sensitivity and specificity. Cells associated with large cell lymphomas tend to have low recovery and lower viability, which compromises diagnostic sensitivity and raises the likelihood of false-negative results. Although CD30 has been detected by flow cytometry with high sensitivity and specificity in lymph nodes of cHL using a complex method of exposing and enriching for sparse tumor cells,36 results are more reliable in fluids, breast implant–associated ALCL, and primary effusion lymphoma. As a consequence of this inconsistency, flow cytometry has not been widely implemented in clinical studies to select patients for CD30-directed therapy. Cell block for CD30 IHC is therefore recommended when screening patients for CD30 in cytology samples.
Although CD30 IHC testing lacks universal standardization, it has the potential to be a reliable diagnostic and predictive method. The overall IHC process is straightforward but involves many steps that must be optimized and consistently implemented. The preanalytic phase for IHC ends with an unstained slide, the analytic phase consists of an IHC protocol and a pathologist's readout of the stained slide, and the postanalytic phase includes the pathologist's interpretation, reporting, and the treating physician's interpretation of reported result(s) in a clinical context.6 In the absence of universally accepted standards and methodology, individual clinical laboratories have developed their own tests (laboratory-developed tests [LDTs]) and acceptability criteria for the assay performance, within the context of applicable regulatory frameworks.
Harmonization of IHC staining results and agreement on acceptability criteria for assay performance are imperative for LDTs to minimize day-to-day variation between instruments, personnel, and institutions. The Nordic Immunohistochemical Quality Control, a Scandinavian proficiency testing program for IHC, provides guidance on how to achieve best signal-to-noise ratios for optimal and comparable CD30 IHC protocol results (including providing highly sensitive and specific IHC protocols as well as recommended tissue controls) on their Web site (www.nordiqc.org) based on 5 proficiency testing runs and the results of 907 participating laboratories that provided detailed descriptions of the IHC protocols used. The most recently posted NordiQC assessments on CD30 (runs 11, 25, 31, 43, and 51) show that 71% to 92% of labs produce sufficient (optimal or good) results. (Note that NordiQC runs are described for reference only and should not be construed as either a recommendation for laboratories to participate in NordiQC or an endorsement of this program.) Testing highly expressed CD30 in cHL or ALCL may be acceptable within labs; however, demonstration of CD30 in low-level antigen-expressing cells is more challenging, making optimal calibration of the IHC assay more critical, but definitely achievable for any clinical IHC laboratory.37
Although vendor-supplied kits and test package insert guidelines may be informative, it is critical that LDT protocols be developed and validated as per existing guidelines and based on the purpose/clinical applications of the IHC assay.38,39 Harmonization of the analytic sensitivity of the protocols as well as the reference materials used for daily quality control are of particular importance and require selection of tissues with known descriptive low limit of detection (eg, CD30+ centroblasts in germinal centers in benign tonsil as recommended by NordiQC; Figure).
![CD30 staining, optimal and insufficient. Tonsil is recommended as a positive control for CD30. Optimal staining highlights activated lymphocytes in follicles, mostly in germinal centers, as well as interfollicularly. Although the number of positive cells can vary greatly, the goal is to demonstrate the range of CD30 expression in germinal centers (A) ranging from weakly positive cells (1+) to strongly positive cells (4+). Some germinal centers may not show a sufficient number of positive cells, particularly when side by side with a germinal center with a large number of CD30+ cells (B). To ensure a representative sample, the tonsillar tissue should always include multiple reasonably large germinal centers. Suboptimal protocols can show cells that are strongly positive for CD30, especially in interfollicular areas (C); germinal centers may show no positive cells (D and E). Here, only the large germinal center is informative (E). Some optimal protocols may show focally positive endothelial cells (F); this is acceptable but not required. Immunohistochemistry, anti-CD30 antibody (clone JCM 182, Leica/Novocastra) (original magnifications ×100 [A, B, and D], ×40 [C], and ×200 [E and F]).](https://allen.silverchair-cdn.com/allen/content_public/journal/aplm/147/1/10.5858_arpa.2021-0270-oa/3/m_i1543-2165-147-1-79-f01.png?Expires=1750159545&Signature=Imo4INqTU5Qsg98l3VyCm5qd53V0T67is-Si~03ejzXhzwLgADAui4SVkptdGgr2y4JnK833pAVtfVJdABzvCn18RX8jK4yUSrnw-oPuthj3mC8hiiPNF~Yh3Caxe8j1zVsRJdllw1cdfnT6vB13h97mbeskoEeRrgt1dTkXeK~ll9vZiKXtba~bk6nk4qqn3JH5AXKC~impMq3CesdX643Q4n7-MY3iVqr9KMJmeqlsSfEQljx1~m72bIIgnhdiryb0qT7erFRsb6f2aEBpURz8boN~ggUOsH~ZKRxfz5UrsrfUTZzEQFd4PzGm32Ei2cn6yOp0Sz2579fwnpffVA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
CD30 staining, optimal and insufficient. Tonsil is recommended as a positive control for CD30. Optimal staining highlights activated lymphocytes in follicles, mostly in germinal centers, as well as interfollicularly. Although the number of positive cells can vary greatly, the goal is to demonstrate the range of CD30 expression in germinal centers (A) ranging from weakly positive cells (1+) to strongly positive cells (4+). Some germinal centers may not show a sufficient number of positive cells, particularly when side by side with a germinal center with a large number of CD30+ cells (B). To ensure a representative sample, the tonsillar tissue should always include multiple reasonably large germinal centers. Suboptimal protocols can show cells that are strongly positive for CD30, especially in interfollicular areas (C); germinal centers may show no positive cells (D and E). Here, only the large germinal center is informative (E). Some optimal protocols may show focally positive endothelial cells (F); this is acceptable but not required. Immunohistochemistry, anti-CD30 antibody (clone JCM 182, Leica/Novocastra) (original magnifications ×100 [A, B, and D], ×40 [C], and ×200 [E and F]).
CD30 staining, optimal and insufficient. Tonsil is recommended as a positive control for CD30. Optimal staining highlights activated lymphocytes in follicles, mostly in germinal centers, as well as interfollicularly. Although the number of positive cells can vary greatly, the goal is to demonstrate the range of CD30 expression in germinal centers (A) ranging from weakly positive cells (1+) to strongly positive cells (4+). Some germinal centers may not show a sufficient number of positive cells, particularly when side by side with a germinal center with a large number of CD30+ cells (B). To ensure a representative sample, the tonsillar tissue should always include multiple reasonably large germinal centers. Suboptimal protocols can show cells that are strongly positive for CD30, especially in interfollicular areas (C); germinal centers may show no positive cells (D and E). Here, only the large germinal center is informative (E). Some optimal protocols may show focally positive endothelial cells (F); this is acceptable but not required. Immunohistochemistry, anti-CD30 antibody (clone JCM 182, Leica/Novocastra) (original magnifications ×100 [A, B, and D], ×40 [C], and ×200 [E and F]).
The current recommendations for different phases of CD30 IHC assay are preanalytic phase, analytic phase, and postanalytic phase.
During the preanalytic phase, tissue should be fixed in a 10% neutral-pH phosphate-buffered formalin for 8 to 72 hours, with a preferred 24-hour maximum.6 Note that if a fixative other than formalin is used for CD30 testing with cell blocks, then internal validation of preanalytic conditions should be performed.
During the analytic phase, IHC protocols and control tissues are followed as per NordiQC recommendations and pathologist's readout (see recommendation 3 for details).
During the postanalytic phase (see recommendation 4 for details), the pathologist's interpretation and reporting for diagnostic purposes, and reporting the CD30 pathologist's readout results for potential use as a predictive biomarker (interpretation for predictive purposes is usually done by hematologists/oncologists), occur.
Several areas remain that require additional research. Although it is currently only possible to use the approach of best signal-to-noise ratio and to rely on descriptive low limit of detection (the presence and the intensity of staining of CD30+ centroblasts in germinal centers), further research should determine the level and the range of analytic sensitivity required for clinically acceptable diagnostic accuracy of the CD30 IHC assay as a predictive biomarker. Furthermore, it is unknown which readout method/scoring is applicable to each disease/diagnosis. It is also incompletely understood if cellular localization of CD30 protein expression or percent positive cells (or both) is important for predicting clinical responses for different lymphomas.
No alternative CD30 testing methods are currently recommended for determination of CD30 status for predictive purposes. For alternative CD30 testing methods to be considered, adequate evidence must be generated by clinical trials that show that such alternative methods are predictive of response to targeted therapy for a specific purpose (disease/drug/diagnosis).
Recommendation 3: Assay Readout
Assuming that the internal and external controls for the IHC protocol performed well, report any limiting issue (eg, crush artifact, necrosis, rare tumor cells) first, and subsequently report CD30 expression based on what is observed.
Given variation of staining intensity in non-Hodgkin lymphoma and the use of nonstandardized IHC protocols, any staining intensity that is recognized as “specific staining” by a pathologist should be considered when estimating percent positive cells.
Because no cutoff has been defined for CD30-targeted agents in PTCL,40 estimate percent positive tumor cells as follows: 0%, 1% to 10%, and in 10% increments or ranges thereafter. This method should capture relevant levels from prior clinical trials so that individual cases can be placed in the context of those trials in PTCL, CTCL, and diffuse large B-cell lymphoma.14,16–19,25,26,41
Record descriptively if nontumor cells (eg, inflammatory background cells) are positive and estimate the percent positive nontumor cells if possible. The clinical relevance of CD30 expression in nonneoplastic cells in the microenvironment is unknown and remains controversial. Given that many studies have demonstrated clinical responses with a CD30-directed agent irrespective of CD30 expression, including tumors with visually undetectable CD30 by IHC,14,16,24–26 more studies are needed to understand this phenomenon. Until research provides evidence supporting clinical activity or lack of activity with CD30-directed therapy targeting infiltrating nonmalignant cells, pathologists are recommended to report what is observed.
In addition, regarding readout, report CD30 expression based on what is observed of any staining intensity (membrane, cytoplasmic, and Golgi-type staining or any combination of these is acceptable). Nuclear staining is not acceptable, and the assay that produced nuclear staining (with or without other patterns) should be considered invalid and the staining repeated. Counting of cells in the lymphoma and microenvironment areas is not recommended. Because of tissue heterogeneity, if the core biopsy is negative and more core tissue is available, it may be helpful to test the rest of the specimen. Excisional biopsy is the preferred specimen type for testing. However, because tissue heterogeneity may also affect excisional biopsies, any negative result may trigger testing of additional paraffin blocks if available. Core needle biopsies with tumor may not contain any positive internal control; therefore, occasional completely negative biopsies may indicate a need for repeat staining of the same block, if no others are available. See the Figure for expected results with controls.
Recommendation 4: Test Interpretation and Reporting
For diagnostic purposes, the interpretation of the CD30 results should follow published diagnostic guidelines.6,42 The pathology report should also include the pathologist's readout of the CD30 IHC staining as per above (see recommendation 3) to facilitate the use of CD30 results as a predictive biomarker in all lymphomas where CD30 is not definitional for the diagnosis. Because CD30 expression results for use as a predictive biomarker are estimated as a percentage of positive tumor cells and no cutoff between “negative” and “positive” results is included in the report, pathologists are not required to provide an interpretation or an assessment of the potential clinical significance of these results.
DISCUSSION
The rationale for developing CD30 testing by IHC and reporting best practices stems from the need for clinicians and researchers to expand a diagnostic biomarker into a predictive biomarker. From the clinician's perspective, previously characterized disease-specific descriptions used for diagnostic purposes are relevant, as well as overall presence of detectable CD30, to inform targeted treatment decisions. Emerging clinical evidence for CD30-directed agents demonstrates improved outcomes for patients with various subtypes of T-cell malignancies.17,18,24,25,43,44
However, standardization and harmonization of the CD30 IHC testing and reporting as a predictive biomarker have been delayed considering that anti-CD30–targeted therapy has already entered clinical practice. Arbitrary thresholds for positivity, which have no reference to a control, are often used. The various LDTs, separately developed by each laboratory, do not have known clinical or even analytic sensitivity. The analytic sensitivity of the CD30 IHC assay is dependent on the specific methodology surrounding the collection of the samples, IHC protocol applied (eg, choice of the primary antibody clone, antigen retrieval, detection system, instrument used, etc) and also (and equally importantly) on the pathologist's readout of the staining results. It is therefore entirely possible that inconsistent and seemingly irreconcilable results in published literature (ie, tumor responses to CD30-directed therapies where biopsies report CD30 is undetectable by IHC) may be secondary to differences in analytic sensitivity and specificity of various LDTs for detection of CD30. Within the last few years, practicing hematopathologists and general pathologists have stated that the minimum percentage of CD30-expressing cells needed for treatment was approximately 20% (courtesy of Seagen, market research, unpublished data). The cutoff thresholds of ≥10% CD30 expression used to define therapeutic interventions in CD30-targeted phase 3 studies were based on regulators' feedback for a defined patient population and opinions of expert clinical investigators and hematopathology advisors, rather than on significant analysis of CD30 expression as a predictive biomarker. Alternatively, phase 1 and phase 2 studies used cutoff thresholds of either any CD30 reported expression or greater than 0%. Herein we sought to identify aspects of CD30 IHC testing and reporting that could be improved and standardized to help inform treatment decisions.
Despite acceptance of CD30 testing as a diagnostic indicator, pathology reports collected as part of a PTCL registry revealed that CD30 was not assessed in 23% of the diagnostic workups for PTCL.45 According to chart reviews conducted in 2017 and 2019, as many as 1 in 3 suspected PTCL cases were not screened for CD30 expression, in part because of a lack of awareness of CD30 as a therapeutic target (courtesy of Seagen, market research, unpublished data). For decades, CD30 had generally been validated in clinical IHC laboratories as a diagnostic biomarker (diagnostic utility), but the reliability of a given positive CD30 threshold to inform targeted therapeutic decisions and predict clinical response is unknown. Tissue variability with biopsy site and type (eg, intratumoral, between lesions) can have effects on CD30 reactivity. Different cellular localization of CD30 (eg, Golgi versus membrane versus cytoplasmic) as well as technical limitations of detecting low levels of expression of cell surface markers with standard IHC represent an additional challenge. Intratumoral heterogeneity has also not been adequately studied outside of CTCL and can potentially alter the results of the CD30 analysis in any type of sample, not only when limited tissue samples are available (eg, needle core biopsies).
The growing interest in CD30 expression as a predictive biomarker for guiding treatment has driven many laboratories to develop their testing and reporting methods from simple positive/negative (“nominal data”) to reporting percent positive cells (“ordinal data”). CD30 testing and reporting methods, however, are inconsistent across laboratories. Although most laboratories can accommodate clinician requests in terms of reported variables and reporting format, the quantity or range of CD30 measured in the tumor or microenvironment that is deterministic of a response to a CD30-targeted agent remains largely unknown. The clinical relevance of CD30 expression in nonneoplastic cells in the microenvironment is also unknown. It is possible, similarly to programmed death ligand-1 (PD-L1) testing in various diseases (PD-L1 IHC 22C3 pharmDx. Instructions for Use. Agilent Technologies Inc), that the combined total score of CD30 expression in tumor cells, lymphocytes, and macrophages relative to all viable tumor cells is clinically relevant. The hypothesis of a bystander effect has been proposed, causing the cytotoxic agent to diffuse from the targeted CD30 reactive cells, resulting in cell death within the tumor microenvironment.46 A study in CTCL evaluated several microenvironmental factors and found that M2‐type tumor‐associated macrophages coexpress CD30. A hypothesis put forth by the authors postulate that CD30-directed agents may target CD30-expressing M2 macrophages, disrupting their tumor-promoting effect and diffusing the microenvironment with a cytotoxic agent to promote tumor cell death.47 Therefore, it is recommended that practicing hematopathologists/pathologists should report what is observed in the tumor and microenvironment without interpretation of the potential significance of the results.
Despite many unknown parameters related to CD30 testing in lymphoma as a predictive biomarker, the assay is currently widely used and readily available. Other methods of CD30 testing, such as flow cytometry, multispectral imaging, RNA scope, or mRNA expression, which could be more sensitive in CD30 detection, are not routinely used when assessing therapeutic interventions. In the absence of a universal protocol, perhaps the best way to increase clinical utility is for the treating physician (usually an oncologist/hematologist) and the pathologist to discuss/audit local CD30 testing practices in the light of published clinical trial results and testing recommendations.
Therefore, we offer expert opinions and recommendations for best practices for transitioning of CD30 testing by IHC from a diagnostic to a predictive biomarker. We aim to harmonize laboratories that are already testing for CD30 for diagnostic purposes with methods and results reporting at the currently achievable levels, without major investment. We propose a framework for testing and reporting CD30 that can be easily implemented and operationalized, despite the limitations in current testing procedures. Lastly, we describe an approach to identify those patients who would benefit from CD30 testing, either diagnostic or therapeutic. We hope that adoption of the proposed recommendations will improve the reproducibility of CD30 assay results between institutions and their protocols as well as between (hemato)pathologists; this would ultimately also enable more reliable testing in clinical trials. Methodology transfer from clinical trials to clinical practice will ultimately increase both the diagnostic accuracy of CD30 testing and its clinical utility for informing therapeutic decisions.
All authors, except Julie Lisano, PharmD, who facilitated discussions, along with Dennis D. Weisenburger, MD (City of Hope Comprehensive Cancer Center, Duarte, California), participated as members of the panel convened to develop general best practice recommendations for reporting CD30 expression. Medical writing assistance was provided by Lauren Angotti (BioBridges) with funding from Seagen Inc.