Multiple Myeloma: Where the Scan Changes the Plan

Staging PET/CT: what to look for

Lesion types on FDG PET/CT [5]

• Focal lesion (FL): a visually detectable focus of FDG uptake above the surrounding marrow, in the skeleton (excluding sites of physiological uptake), on 2 or more adjacent images, with or without osteolysis on CT.
• Diffuse marrow involvement: diffuse uptake in the axial skeleton above liver, homogeneous or heterogeneous, which may extend to the peripheral skeleton.
• Paramedullary disease (PMD): soft tissue extending from, and contiguous with, involved bone.
• Extramedullary disease (EMD): abnormal soft tissue with no contiguous bone involvement.
• Osteolytic lesions of 5 mm or larger on the CT component [2][5].

What to expect at diagnosis

• FDG PET/CT detects myeloma lesions with a sensitivity of about 90% and a specificity of 70–100% [5].
• In two prospective multicentre cohorts of newly diagnosed patients (134 and 268 patients), PET/CT was normal in 9% and 20%, and EMD was present in 10% and 8%. Paramedullary disease was present in 18% of the larger cohort [5].
• Up to 80% of patients present with osteolytic bone lesions at diagnosis [13].
• FDG PET/CT assesses bone damage with relatively high sensitivity and specificity, detects extramedullary sites of proliferating plasma cells, and adds prognostic information [3].

Field of view [5]

• Image from the skull vertex to the toes, or at least to the knees, with the arms alongside the body; PET coverage should match the CT field of view [5].

Typical spread and distribution

Skeletal distribution. Involvement follows the red marrow of the axial skeleton and proximal long bones. The site frequencies in the chart below come from a radiographic series of 237 patients (Ferrara, 1984–1990) [17].

Extramedullary disease rates. True EMD means soft-tissue plasmacytomas that arise by hematogenous spread with no contact with bone. Paraskeletal plasmacytomas, which grow from bone, are a separate category [15]. Definitions and diagnostic methods vary between studies [15]. In a single-institution series of 1,304 patients (1970–2018), paraskeletal plasmacytomas were present in 17.6% of patients at diagnosis and in 14.6% of 967 patients at first relapse, so the extramedullary share rose while paraskeletal disease did not [16]. EMD incidence and plasmacytoma rates are given in the chart below.

Common EMD sites.

• At diagnosis: skin and soft tissues [15].
• At relapse: liver, kidneys, lymph nodes, CNS, breast, pleura and pericardium [15].
• CNS involvement is rare but carries an even worse prognosis than EMD elsewhere, particularly with leptomeningeal disease [15].

Typical sites of spread

% of patients in each cited series, on a 0–100% scale. Solid bars are single values; hatched bands show the range reported across studies.

Where PET/CT changes stage or management

• Bone CRAB criterion on the CT component. One or more osteolytic lesions of 5 mm or larger on the CT component of PET/CT fulfils the bone ("B") CRAB criterion [2][5].

Changes management: start anti-myeloma therapy instead of observation, because a lytic lesion of 5 mm or more on the CT component is a criterion for starting therapy [2][5].

• FDG-avid focal lesion without osteolysis. Under the IMWG diagnostic criteria, underlying osteolysis is required before a positive FL on PET/CT counts as a criterion for starting therapy; FLs without osteolysis are not yet myeloma-defining events [5]. In smoldering myeloma (SMM), such uptake still carries prognostic weight: in 122 patients, 2-year progression to symptomatic myeloma without therapy was 75% with a positive PET/CT (with or without osteolysis) and 30% with a negative scan. In a prospective study of 120 patients, 2-year progression was 58% with a positive PET/CT (all without osteolysis) and 33% with a negative scan [5].

Changes management: the patient stays on observation as higher-risk SMM rather than starting treatment, with yearly imaging for at least 5 years [5].

• Solitary plasmacytoma vs myeloma. FDG PET/CT is first-line imaging for suspected extramedullary solitary plasmacytoma [5][12]. In presumed solitary bone plasmacytoma, FDG PET may show additional lesions with therapeutic implications in 33–55% of patients [12]. Two or more FLs on the initial PET/CT were associated with a shorter time to progression to myeloma [5].

Changes management: localized radiation therapy to the single lesion (40–50 Gy), only once PET/CT has excluded additional lesions; additional lesions rule out solitary plasmacytoma and have therapeutic implications [12].

• Extramedullary disease (EMD). PET/CT detects extramedullary sites of proliferating plasma cells [3]. EMD carries a poor prognosis, considerably worse than myeloma without soft-tissue plasmacytomas [15]. Paramedullary disease was independently associated with shorter progression-free survival [5].

Changes management: in transplant-eligible patients, an intensive regimen with stem-cell transplant is a theoretical option only, as prospective data are lacking; reassess EMD on PET/CT 3 months after treatment starts [15].

• Prognostic PET features. In a prospective series of 192 newly diagnosed patients, at least 3 focal lesions, SUVmax above 4.2 and EMD each predicted shorter progression-free survival. SUVmax above 4.2 and EMD also predicted shorter overall survival, and in multivariable analysis SUVmax above 4.2 and EMD at baseline were independent adverse factors for progression-free survival [8]. In a harmonized analysis of 227 newly diagnosed patients from two prospective phase III trials, baseline bone SUVmax remained an independent predictor of overall survival in a multivariable model that included the R-ISS [9][14], with an overall-survival threshold of 7.1 [9].

Changes management: no specific treatment change is established; these features mark higher baseline risk only [8][9].

What to report

Before reading, note [20]

• Indication: staging, response assessment or suspected relapse; first line or relapse.
• Current treatment; prior autologous or allogeneic transplant; prior radiotherapy or surgery, including vertebroplasty.
• Recent G-CSF or steroid therapy, and MRD status if known.
• Symptomatic sites, including any clinical concern for cord or nerve-root compression.

Technique [5][20]

• Injection time and administered activity, field of view (vertex to toes, or at least to knees, arms down), comparison study, image quality and limitations.

Findings at staging (report by IMPeTUs) [5]

• Marrow: diffuse uptake relative to liver (Deauville score), and whether the ribs and limbs are hypermetabolic.
• FLs: number and site, with or without osteolysis on CT; Deauville score and SUVmax of the hottest lesion.
• Lytic lesions on CT, counted by the same groups.
• Paramedullary disease: present or absent.
• EMD: present or absent, with site.
• Fractures on CT.

Findings during or after therapy [4][5]

• Comparison with the baseline PET/CT.
• Deauville score in the marrow and in each previously involved FL, paramedullary site and EMD site.
• New FLs, paramedullary disease or EMD.
• CT component: new osteolytic lesions or growth of known lesions.
• IMPeTUs response category, written in full (complete, partial, stable or progressive metabolic response or disease).

Impression [5][20]

• A clear, concise conclusion, with any recommendation for follow-up or further tests.
• Check the images before the patient leaves, and call the referring team directly for urgent findings such as fractures or spinal cord compression [5].

Response and restaging

• The IMWG considers FDG PET/CT the preferred imaging technique for evaluating and monitoring response to therapy [3][5]. PET/CT separates metabolically active lesions from inactive residual osteolytic lesions [5].
• There is no expert consensus on the best timing of post-treatment PET/CT; interim (after induction), end-of-therapy (before maintenance) and maintenance-phase scans are all used [5]. Normalization can be seen as early as day 7 post-induction, and the normalization rate rises through induction, transplant and maintenance [5].

Deauville 5-point scale as applied in myeloma [5]

• Score 1: no uptake.
• Score 2: uptake at or below mediastinal blood pool.
• Score 3: uptake above mediastinal blood pool and at or below liver.
• Score 4: uptake above liver (at least 10% more).
• Score 5: uptake markedly above liver (at least twice).
• Scoring is visual. Target-to-background ratios may support it, using a region of interest of 3 cm radius or more in the central liver and a second region within the aortic lumen, such as the arch, for blood pool. The criteria were validated on OSEM reconstructions, so time-of-flight or other noise-optimizing algorithms should not be used when applying them [5].

IMPeTUs criteria

• IMPeTUs scores diffuse marrow uptake, the hottest focal lesion and EMD on the Deauville scale. Diffuse marrow uptake is flagged "A" when the limbs and ribs are hypermetabolic. FLs are recorded by site (skull, spine, extra-spinal), and focal and lytic lesions are counted as none, 1–3, 4–10 or more than 10, along with paramedullary disease and fractures [5][6]. In 86 patients (211 scans), IMPeTUs was highly reproducible, with the highest agreement for Deauville score 4 [7].
• Response categories (EANM 2024, from IMPeTUs) [5][10]:
  • Complete metabolic response: uptake at or below liver (Deauville 1–3) in the marrow and in previously involved FLs, paramedullary disease and EMD.
  • Partial metabolic response: fewer or less avid marrow sites, FLs, paramedullary disease or EMD, but with persistent lesions above liver (Deauville 4 or 5).
  • Stable metabolic disease: no significant change from baseline.
  • Progressive metabolic disease: new FLs, paramedullary disease or EMD compared with baseline, consistent with myeloma.
• In a joint analysis of 228 transplant-eligible patients from two phase III trials, marrow and focal-lesion uptake below liver (Deauville score below 4) before maintenance independently predicted longer progression-free and overall survival. This is proposed as the standardized definition of PET complete metabolic response [10]. The EANM recommends these categories pending external validation before formal adoption into the IMWG response criteria [5].

IMWG imaging criteria (Kumar 2016)

• Imaging plus MRD-negative. Requires bone marrow MRD negativity by next-generation flow or sequencing, plus disappearance of every area of increased tracer uptake seen at baseline or on a preceding PET/CT, or a decrease to less than mediastinal blood pool SUV or less than surrounding normal tissue [4][5]. This threshold is blood pool, not the liver threshold used by IMPeTUs.
• Imaging progression. New osteolytic lesions, a 50% or greater increase from nadir in the sum of the products of the maximal perpendicular diameters of more than one lesion, or a 50% or greater increase in the longest diameter of a previous lesion more than 1 cm in short axis [4][5].
• Unless there is disease progression, the IMWG does not recommend changing treatment on post-treatment imaging alone [5].

Prognosis after therapy

• Three months after autologous transplant, PET/CT was negative in 65% of patients in the Zamagni series. Those patients had superior 4-year progression-free and overall survival, and persistent FDG uptake after transplant was an independent adverse factor for progression-free survival [8]. Persistent FLs with SUVmax above 4.2 after induction and after transplant were associated with shorter time to progression [5].
• MRD-negative patients with a positive post-treatment PET/CT are uncommon (5–7%) but may be at higher risk of early relapse; negative MRD with a negative post-treatment PET/CT is associated with longer progression-free survival [5].

Changes management: residual uptake alone does not change treatment unless there is progression; it changes follow-up to yearly FDG PET/CT [5].

Surveillance and suspected relapse

• Myeloma. The IMWG recommends yearly FDG PET/CT follow-up for patients whose post-treatment PET/CT is positive, since they are at higher risk of early progression [5]. Where MRD status is available, EHA-ESMO guidance recommends yearly FDG PET/CT in marrow MRD-negative patients to confirm extramedullary MRD negativity [5]. The same imaging technique should be used at each follow-up stage to allow comparison [5].
• Solitary plasmacytoma. Because of the risk of progression to myeloma (60% for bone and 20% for extramedullary plasmacytoma within 3 years), the IMWG recommends yearly imaging for the first 5 years, using the same technique as at diagnosis [5].
• Suspected progression or relapse. A new plasmacytoma or osteolytic lesion is one of the criteria defining relapse [4][5]. FDG PET/CT helps separate active from inactive lesions and detects EMD, which is more common at relapse and shortens both time to progression and overall survival [5]. At relapse, FL number, EMD and SUVmax are associated with progression-free survival, though more data are needed to validate them [5].

Changes management: a change of treatment, because a new osteolytic lesion or plasmacytoma is a relapse criterion, and IMWG allows imaging alone to change treatment only for progression [4][5].

Pearls and pitfalls

• Diffuse vs focal marrow uptake. Anemia, common in myeloma, can raise marrow uptake diffusely as a compensatory response. That lowers contrast for detecting coexisting focal lesions and is hard to tell from diffuse infiltration. Marrow uptake of Deauville 4 or higher, and marrow involvement in the limbs, make true infiltration more likely [5]. At baseline, every patient in the two-trial cohort had diffuse marrow uptake (35.5% at Deauville 4 or higher). Before maintenance, 98% still showed some diffuse uptake, mostly at Deauville 2–3 [10]. The differential for diffuse marrow infiltration includes mastocytosis, myelofibrosis, lymphoma and leukemia [20].
• Read early focal uptake carefully. Without osteolysis on CT, focal marrow uptake is difficult to interpret; it is good practice to require focal uptake on two or more adjacent images [5].
• Reading the CT component. On CT, endosteal thinning suggests tumour, while poorly defined, inconspicuous areas favour red marrow [20]. In untreated disease, a mixed bone-lesion pattern or a sclerotic rim around a lytic lesion argues against myeloma; early pedicle involvement and sclerotic lesions favour metastasis, while involvement of the disk and adjacent vertebrae favours myeloma or plasmacytoma [20].
• Post-treatment changes. PET/CT distinguishes metabolically active from inactive disease, so a residual lytic lesion on CT does not by itself mean active myeloma [3]. The IMWG recommends a baseline PET/CT for comparison. If the baseline scan is FDG-negative, post-treatment PET/CT may not be suitable for assessing metabolic response [5].
• G-CSF and marrow stimulation. Growth factors (G-CSF and GM-CSF) increase marrow FDG uptake, and the effect generally lasts more than 2 weeks after the last dose [18]. FDG PET/CT should be avoided in the days after G-CSF; if it cannot be, interpret diffuse marrow abnormalities with caution [20]. Marrow hyperplasia from anemia or cytokine therapy is a recognised cause of false-positive FDG uptake [19].
• FDG-negative myeloma. About 10–20% of newly diagnosed patients have no FDG uptake at staging despite proven marrow infiltration or lytic lesions on CT [5]. In 227 newly diagnosed patients, 11% were FDG PET false-negative, and those cases had markedly lower hexokinase-2 expression (5.3-fold) [11]. A normal baseline PET/CT is still associated with better outcomes than a positive one [5].
• Other pitfalls.
  • Recent fractures can be FDG-avid for at least a month [5].
  • Osteosynthetic hardware can cause false-positive uptake at the bone interface [5].
  • Degenerative change, such as recent Schmorl nodes, and inflammatory uptake in tendons, osteophytes and arthrosis can mimic lesions; the low-dose CT images help [5].
  • Inflammation can mimic EMD; use targeted imaging or biopsy when in doubt [5].
  • T-cell-redirecting bispecific antibodies can cause immune-related findings [5].
  • Recent high-dose steroid therapy is a recognised cause of false-negative FDG studies [19].
  • Brown-fat uptake can be reduced by keeping the patient warm before injection [5][18].

Key Images

‹

›