Photon Counting Detector Computed Tomography: A New Frontier of Myeloma Bone Disease Evaluation Open Access

Computed tomography (CT) is critical when evaluating plasma cell disorders to identify osteolytic lesions which confirm the diagnosis [1]. Conventional whole-body low-dose CT (WBLDCT) is the primary choice for myeloma imaging; however, fundamental limitations in the energy-integrating detectors (EIDs) used in nearly all of the current clinical CT scanners hinder the ability to resolve small lytic lesions and focal intramedullary lesions, both characteristic features of multiple myeloma [2]. Additionally, WBLDCT is acquired at a low dose and consequently, exhibits higher image noise [3]. The compounding effect of lower spatial resolution and high image noise results in CT images that may be of suboptimal diagnostic quality for this imaging task. With advanced risk stratification of multiple myeloma and new definitions of ultra-high-risk disease groups, there is a need for a more sensitive CT imaging techniques [4].

Photon counting detectors (PCDs) represent a fundamental shift in CT detector technology. These detectors directly convert individual X-ray photons to electric signal whereas the EIDs used in current clinical CTs employ a two-step process where X-rays are first converted to visible light and subsequently converted to electric signal [5, 6]. With PCDs’ direct conversion approach to X-ray detection, information about the energy of each photon is recorded based on the magnitude of the electrical signal generated from photon events. An important implication of this is that with PCDs, photons that carry information from the patient can be selectively used for image creation while those photons characteristic of electronic noise can be excluded [5]. This advancement in CT technology led to the clinical approval of a PCD-CT by the US Food and Drugs Administration in 2021, the first announcement of its kind since 2013 [7].

The impact of PCD-CT on the clinical care of multiple myeloma patients has not yet been established. The following cases were selected to highlight the superior lesion detection and depiction of intramedullary disease compared with current clinical standard WBLDCT.

We describe 2 patients with multiple myeloma who underwent a clinically indicated WBLDCT as part of their work up and were enrolled in a research study with PCD-CT [8]. The electronic medical records were reviewed to abstract the corresponding clinical course and management. The prospective study was approved by the Institutional Review Board and was conducted in accordance with the Declaration of Helsinki.

An 81-year-old man was being followed for IgG kappa smoldering myeloma. At his diagnosis 3 years prior, M spike was 2.3 g/dL, kappa free light chain (FLC) was 10 mg/dL, lambda FLC was 1.38 mg/dL (ratio 7), hemoglobin was 14.6 g/dL, creatinine was 0.91 mg/dL. Bone marrow at diagnosis demonstrated 10% kappa light chain restricted plasma cells, standard risk fluorescence in situ hybridization and the baseline imaging evaluation included a WBLDCT and a PET scan which showed an L1 compression fracture but no lytic lesions nor FDG avid metabolic activity. Given the diagnosis of intermediate risk smoldering myeloma by 20/2/20 criteria [9], he was monitored every 3 months clinically and with laboratories, as well as annual imaging with PET scan for the first 2 years, then CT skeletal survey. Over time, there was slow increment in his FLCs and M spike and at the time of his reevaluation with PCD-CT 3 years after diagnosis, his kappa FLCs had risen to 31.6 mg/dL, lambda FLC was 0.91 mg/dL (ratio 34.4), and M spike was 3.5 g/dL. His hemoglobin was 13.7 g/dL, which was about a 1 g drop from his usual baseline and the patient had reported significant fatigue. A WBLDCT acquired on EID-CT scanner demonstrated no imaging stigmata of active myeloma. The patient also underwent WBLDCT on a PCD-CT scanner on the same day. Images from this scan revealed a lytic lesion in a thoracic vertebral body and a focal lesion in the sacrum (Fig. 1). These were occult on the clinical EID-CT scan. He subsequently had bone marrow evaluation which reported 30–40% kappa light chain restricted plasmacytosis, and a cervical MRI was reviewed which also demonstrated marrow-replacing lesions consistent with multiple myeloma. The patient was subsequently started treatment for multiple myeloma.

A 67-year-old woman with IgA kappa multiple myeloma had received 5 prior lines of therapy over 11 years. At diagnosis IgA was 1,140 mg/dL, there was no measurable M spike, kappa FLC was 79.3 mg/dL, lambda FLC was 0.24 mg/dL (ratio 330). Bone marrow at diagnosis demonstrated 55% kappa restricted plasma cell involvement. There were no high risk abnormalities on fluorescence in situ hybridization and her baseline MRI had revealed multiple lytic lesions. Her last line of therapy was a B-cell maturation antigen (BCMA) antibody-drug conjugate for only 3 months, stopped due to treatment-related side effects. By that time, the patient had achieved a hematologic complete response (kappa and lambda FLC below detection and no monoclonal protein on the matrix-assisted laser desorption/ionization-time of flight mass spectrometry immunofixation [MALDI-TOF Mass-fix]) and she elected to be subsequently monitored off therapy. Her imaging surveillance thereafter consisted of alternating PET-CT and CT skeletal survey every 6 months. At a routine follow-up visit, 11 years after diagnosis and 2.5 years of monitoring after the BCMA antibody-drug conjugate was discontinued, there was evidence of biochemical progression; kappa FLC was 12.4 mg/dL, lambda FLC was 1.45 mg/dL (ratio 8.55), there was no M spike, but there was monoclonal kappa evident on the MALDI-TOF Mass-fix. The conventional EID WBLDCT at that time re-identified the same lytic lesions with no additional or enlarging lesions. However, the margins of these lesions were more clearly delineated on the PCD-CT scan acquired on the same day (Fig. 2). With rising light chains over the 2 subsequent follow-up visits, a new line of myeloma therapy was subsequently offered to the patient.

These cases describe the potential application of PCD-CT for multiple myeloma diagnosis and highlight the impact of this technology. In the first case of smoldering multiple myeloma, PCD-CT enabled the detection of lytic lesions and focal intramedullary lesions which were not obvious on the clinical EID WBLDCT. This is a crucial finding since the presence of myeloma defining bone lesions upstages smoldering multiple myeloma to active disease [4]. The findings on the PCD-CT images more accurately complement the biochemical progression of disease whereas reliance on the EID-CT alone underestimates the disease activity.

WBLDCT is acquired at a fraction of the dose of a standard whole-body CT and is adequate for screening of monoclonal gammopathies [10]. However, further reducing the dose increases image noise and compromises image quality, an effect which is even more pronounced in conventional WBLDCT [11]. PCDs overcome this limitation in two ways. First, PCD pixels are small, such that whole-body imaging can be performed at high spatial resolution so small structures are clearly depicted. Whole-body high-resolution imaging is not feasible with EID systems due to their inherent radiation dose inefficiency. Second, energy thresholding in PCDs, preferentially selects X-ray photons with a high likelihood of conveying anatomic information to generate the CT image while excluding photons likely associated with electronic noise [5, 12]. The combination of these features results in a low-dose CT acquisition with up to 47% lower image noise or improved spatial resolution relative to what is achievable with EID-CT systems [5, 12]. This benefit of PCD is demonstrated in the second index case where small lytic lesions were more readily seen in the pelvis. This is an area that is typically subject to increased image artifact due to the highly dense pelvic bones.

Whole-body MRI with diffusion-weighted imaging remains the gold standard to evaluate the skeleton and marrow infiltration in plasma cell disorders, while FDG-PET-CT is the recommended imaging modality to assess disease activity and monitor subsequent treatment response [1]. There is an important distinction between focal lesions and osteolytic lesions, which may be obscured in clinical practice, and it is important for the practicing hematologist to recognize the difference between the two, as the former does not require destruction of the cortical bone. EID-CT imaging is an excellent modality for evaluating for osteolytic disease, but PCD-CT has superior capacity to image the intramedullary cavity. Though PCD-CT has not, at this time, supplanted MRI or FDG-PET/CT, it is a novel imaging modality, with a potential established role for bone disease assessment in plasma cell disorders as imaging algorithms evolve.

Accurate and sensitive imaging is critical in the diagnosis of smoldering multiple myeloma, an asymptomatic intermediary in the spectrum of monoclonal gammopathies, for which there should not be more than one 5 mm focal lesion present [1]. This can be missed on EID-CT and so if negative findings on CT, the recommendations are to move onto MRI to definitively exclude smoldering myeloma [1]. Because PCD-CT has high-resolution imaging which can better demonstrate intramedullary lesions compared to EID-CT, PCD-CT could add significant value in the diagnostic paradigms for monoclonal gammopathies, including smoldering multiple myeloma.

Oligosecretory or nonsecretory multiple myeloma is another clinical context in which imaging is critical for evaluation and monitoring [13]. Since monoclonal protein blood and urine biomarkers do not meet criteria for measurable disease in these conditions, FDG-PET/CT or MRI imaging is heavily relied upon for assessment and monitoring of disease [4, 13, 14]. This is yet another scenario where PCD-CT could have an established role in diagnosis and management.

Though PCD-CT is a new technology, it offers objective benefits in relation to current gold standard imaging. MRI and PET/CT may be readily available in academic centers, but the same may not apply to community practices given the associated cost. CT imaging is much more accessible in clinical practice and we anticipate that PCD-CT could be feasibly adopted into routine care. Unlike MRI, PCD-CT is fast, completed in a few minutes, and not subject to the motion artifact of MRI [15]. The effective radiation dose of FDG-PET/CT is up to 4–6 times higher than that of PCD-CT [16, 17]. The future potential for PCD-CT will expand to measurement of bone mineral density, and noninvasive monitoring of treatment response through multi-energy capabilities such as virtual non-calcium images without cortical and trabecular bone to better visualize the medullary cavity [5, 18]. Though still investigational, these additional metrics would have considerable impact on assessment of bone health and disease monitoring in myeloma patients.

Just as the therapeutic options for multiple myeloma have rapidly evolved to improve the outcomes for patients, so too have there been parallel advancements in imaging technologies. The superior spatial resolution, lower image noise, and better dose efficiency of PCD-CT ushers in a new era of sensitive CT evaluation of myeloma bone disease. The high-resolution images from PCD-CT have potential clinical value in managing multiple myeloma and its precursor states.

The prospective study was approved by the Institutional Review Board at Mayo Clinic and was conducted in accordance with the Declaration of Helsinki. The IRB number/title is 16-001988 “Potential benefit of Photon-counting CT in Human Subjects.” The approval date was April 5, 2016. Written informed consent was obtained from the patients for publication of the details of their medical case and any accompanying images. There is no information revealing subject identity.

The authors have no relevant conflicts of interest to declare.

The authors acknowledge the support of the Mayo Clinic CT Clinical Innovation Center, and Siemens Healthineers, who own the CT system under the terms of a sponsored research agreement with the Mayo Clinic.

The authors confirm contribution to the paper as follows: study conception and design J.C., K.R., A.F., P.D., S.K., and F.B. Draft manuscript preparation: J.C., K.R., A.F., S.K., and F.B. All authors reviewed the results and approved the final version of the manuscript.

Additional data are not publicly available as it contains propriety information. Further inquiries can be directed to the corresponding author (J.M.C).