Abstract
Purpose
To explore the correlation between semi-quantitative parameters of 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) scans findings and the clinical features of patients with acute leukemia (AL), as well as to evaluate the clinical utility of 18F-FDG PET/CT in the management of AL.
Methods
A retrospective study was conducted with 44 patients newly diagnosed with acute leukemia (AL) at Zhongnan Hospital of Wuhan University between January 2019 and August 2024.
Results
Multivariate analysis revealed that age at diagnosis of AL (odds ratio [OR]: 0.888, P < 0.01) and percentage of blasts in the peripheral blood (PB) (OR: 1.061, P < 0.05) were independent predictors of the appearance of active extramedullary disease (EMD). Kaplan–Meier survival analysis for patients with EMD(+) indicated that those with organ infiltration beyond the lymph nodes experienced markedly reduced overall survival (OS) compared to those without such infiltration (157 days and 806 days, respectively). Furthermore, in the AL subgroup with EMD, the ratio of the maximum standardized uptake value (SUVmax) in the bone marrow (BM) to SUVmax of the liver emerged as an independent prognostic factor for OS (Hazard ratio [HR]: 2.372; 95% confidence interval [CI]: 1.079–5.214, P < 0.05).
Conclusion
18F-FDG PET/CT offers the benefits of being non-invasive and highly sensitive for the thorough evaluation of disease status in patients newly diagnosed with AL. Furthermore, the SUVmax BM/liver ratio is of significant clinical importance for prognosticating outcomes in patients with AL presenting EMD.
Supplementary Information
The online version contains supplementary material available at 10.1007/s13402-024-00993-z.
Keywords: Acute leukemia, Extramedullary disease, PET/CT, Metabolic parameters
Introduction
Leukemia, a malignant disorder of the hematologic system, originates from hematopoietic stem cells and its incidence is among the top ten of all neoplasms worldwide. Leukemia accounts for approximately 474,519 new cases per year, representing 2.5% of all cancer diagnoses and 3.1% of total cancer-related deaths [1]. Acute leukemia (AL) is subdivided into acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML), depending on the cell type from which the leukemia originates. The rapid multiplication of malignant cells in AL severely inhibits normal bone marrow (BM) hematopoiesis and often leads to the invasion of non-hematopoietic tissues and organs, including the skin, liver, and gums [2, 3]. Extramedullary disease (EMD) in AL denotes lesions where leukemia cells infiltrate anatomical sites beyond the bone marrow [4]. Extramedullary involvement has also been increasingly reported in patients with AML and ALL, however, clinical reliance on physical examinations alone or incidental findings from standard imaging procedures may underestimate the true incidence of EMD [5].
18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) is considered effective in evaluating metabolically active EMD in soft tissues and identifying focal lesions in the BM [6]. This innovative imaging modality combines functional metabolic imaging with anatomical structural imaging, representing a significant advancement in the field [7]. While 18F-FDG PET/CT has gained widespread application in the diagnosis, staging, and evaluation of the response to treatment in hematologic malignancies, including lymphoma and multiple myeloma (MM) [8–11]. However, it should be noted that PET/CT has not achieved the same level of utilization in AL as in lymphoma and MM. And the prognostic significance of EMD in AL debated. Some studies indicate that EMD is associated with poorer outcomes; however, others contest its status as an independent prognostic factor for AL [12, 13]. To further elucidate the role of 18F-FDG PET/CT in AL and the impact of EMD on AL prognosis, we performed a retrospective study involving 44 patients. This study aimed to evaluate the clinical utility of 18F-FDG PET/CT by analyzing PET/CT scan indices alongside clinical indicators in patients with AL.
Materials and methods
Patients
In this retrospective analysis, we reviewed 44 patients with an initial diagnosis of AL who were admitted to Zhongnan Hospital of Wuhan University between January 2019 and August 2024 with confirmed diagnosis by histopathology. Inclusion criteria consisted of patients with an initial diagnosis who underwent BM aspiration biopsy, confirmed as AL using clinical data, and those who received a total body 18F-FDG PET/CT scan before starting first-line chemotherapy with satisfactory image quality. Exclusion criteria included patients who had undergone radiotherapy or chemotherapy prior to 18F-FDG PET/CT, those with a history of other malignant tumors, those without standard treatment, those currently pregnant or breastfeeding, and those with significant infections or inflammation.
The study adhered to the principles of the Declaration of Helsinki and received approval from the Ethics Committee of the Zhongnan Hospital of Wuhan University [grant number 2024087 K]. All participants gave their informed consent and underwent 18F-FDG PET/CT imaging.
PET/CT examination
18F-FDG was obtained from Wuhan HTA Pharmaceutical Co., Ltd., Wuhan, China, with a radiochemical purity exceeding 95%. Before imaging, each patient was instructed to fast for 4–6 h, maintain fasting blood glucose levels below 11.1 mmol/L, and receive an intravenous injection of the imaging agent18F-FDG based on body weight (3.7 MBq/kg). After a quiet rest period of 45–60 min, the patients were subjected to PET/CT imaging. Siemens Biograph mCT PET/CT was used for all imaging studies. The CT scanning parameters included a tube voltage of 120 kV, an effective tube current ranging from 40 to 120 mA, an automatic adjustment of mA, and a slice thickness of 3 mm. PET data were iteratively reconstructed using CT data for attenuation correction (2 iterations and 21 subsets), employing the TrueX + TOF (ultra-high-definition-PET) reconstruction method with a Gaussian filter, a magnification set at 1.0, and a full-width at half-maximum (FWHM) of 3.0 mm. During the scan, the patients were positioned with their hands on either side of their body, palms down, and their heads secured in a head frame. The scanning range extended from the roof of the skull to the upper end of the femur, with an acquisition time of 1.5 min per bed. Subsequently, the PET/CT images were co-aligned and displayed using SyngoTrueD software from Siemens Healthcare.
Two experienced nuclear medicine physicians, well-acquainted with the medical histories of each patient, performed a comprehensive analysis of the imaging data. All images were processed, and metabolic indices were quantified using the MEMRS-NM post-processing workstation. The regions of interest (ROI) were identified and delineated, noting the location and number of EMD. Metabolic parameters were measured, including maximum, mean, and peak standardized uptake values (SUVmax, SUVmean, SUVpeak), metabolic tumor volume (MTV), and total lesion glycolysis (TLG), with TLG calculated as SUVmean multiplied by MTV. Standardized uptake values (SUV) for the lumbar vertebrae 3/4 (L3/4) were measured on the sagittal images, defining the bone marrow SUV as BM SUVmax, BM SUVmean, and BM SUVpeak. Similarly, SUVs for hypermetabolic focal lesion (FL) in the bone marrow were recorded as FL SUVmax, FL SUVmean, and FL SUVpeak, respectively. The uptake of FDG in the bone marrow was considered positive if it exceeded that of the liver and negative if it was lower. Furthermore, for each patient, we calculated the SUVmax BM/liver ratio and the SUVmax FL/liver ratio.
Parameters analyzed
At the time of the 18FDG PET/CT scans, all patients were newly diagnosed with AL and had not received prior treatment. The characteristics of the patients, including sex, age at diagnosis, blood counts, and liver and kidney functions, were extracted from medical records. The percentage of BM blasts was assessed from the results of BM aspiration, while the percentage of peripheral blood (PB) blast was determined from PB smears. Laboratory parameters were defined as follows: elevated white blood cells (WBC ≥ 10 g/L), elevated β2-microglobulin (β2M ≥ 3.5 mg/L), and elevated lactate dehydrogenase (LDH ≥ 250 U/L).
Statistical analysis
We utilized SPSS version 25.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 8 software (GraphPad Inc., La Jolla, CA) for statistical analysis. The Shapiro-Wilk test was employed to assess the normality of continuous variables and continuous variables are presented as median with interquartile range followed by either the Mann-Whitney U test for comparing groups. Differences in characteristics between groups were analyzed using the χ^2 test. The Kaplan-Meier method estimated overall survival (OS), which was compared across groups using the log-rank test. OS was calculated from the diagnosis date until death from any cause. Multifactorial analysis of outcome variables was conducted using binary logistic regression, while univariate and multivariate analyses of prognostic factors utilized Cox regression. Receiver operating characteristic (ROC) curves and the area under the curve (AUC) were employed to evaluate diagnostic efficacy. A p-value of less than 0.05 was considered statistically significant.
Results
Clinical characteristics of patients
The study included 44 patients with AL, comprising 20 males and 24 females, with a median age of 54 years (range 27–64.5). Patients were followed until death or until 30 August 2024. The median duration of follow-up was 14.1 months (range 6.58–15.98 months). All patients were subjected to 18F-FDG PET/CT scanning. The clinical characteristics of the patients were documented throughout the follow-up period (Table 1).
Table 1.
Baseline characteristics of patients
| Variable | Total (n = 44) | Patients with EMD (n = 25) | ||
|---|---|---|---|---|
| No. | % | No. | % | |
| Sex | ||||
| Male | 20 | 45.5 | 13 | 52 |
| Female | 24 | 54.5 | 12 | 48 |
| Age | ||||
| ≤ 55 | 24 | 54.5 | 19 | 76 |
| > 55 | 20 | 45.5 | 6 | 24 |
| Disease classification | ||||
| AML | 18 | 40.9 | 14 | 56 |
| ALL | 26 | 59.1 | 11 | 44 |
| Extramedullary symptoms are the first manifestation | ||||
| Yes | 5 | 11.4 | 5 | 20 |
| No | 39 | 88.6 | 20 | 80 |
| WBC (×109) | ||||
| < 10 | 34 | 77.3 | 17 | 68 |
| ≥ 10 | 10 | 22.7 | 8 | 32 |
| β2M (mg/L) | ||||
| < 3.5 | 28 | 63.6 | 23 | 92 |
| ≥ 3.5 | 11 | 25 | ||
| NA | 5 | 11.4 | 2 | 8 |
| LDH (U/L) | ||||
| < 250 | 17 | 38.6 | 10 | 40 |
| ≥ 250 | 26 | 59.1 | 15 | 60 |
| NA | 1 | 2.3 | ||
| Cytogenetics | ||||
| Normal karyotype | 11 | 25 | 7 | 28 |
| Chromosomal abnormality ≤ 2 | 17 | 38.6 | 7 | 28 |
| Complex karyotype | 5 | 11.4 | 4 | 16 |
| NA | 11 | 25 | 7 | 28 |
| High-risk genes | ||||
| Negative | 8 | 18.2 | 5 | 20 |
| Positive | 31 | 70.5 | 18 | 72 |
| NA | 5 | 11.4 | 2 | 8 |
| Complete Remission | 24 | 54.5 | 11 | 44 |
WBC white blood cells, β2M beta-2-microglobulin, LDH lactate dehydrogenase, EMD extramedullary disease, AML acute myeloid leukemia, ALL acute lymphoblastic leukemia, NA not available
Independent risk factors for the development of EMD in patients with AL
Univariate and multivariate analysis of risk factors for EMD
The patients were classified according to the presence of extramedullary lesions into two groups: 25 patients with EMD and 19 patients without EMD. There were no statistically significant differences in sex, normal karyotype, chromosomal abnormality ≤ 2, complex karyotype, presence of high-risk genes, TP53 gene, BM SUVmax, BM SUVmean, BM SUVpeak, and the SUVmax BM/liver ratio between the two groups of patients with AL (P > 0.05). However, significant differences were observed in age at the initial AL diagnosis, WBC counts, BM blast percentage, PB blast percentage, FL SUVmax, FL SUVmean, FL SUVpeak, and the SUVmax FL/liver ratio between the groups (Supplementary Tables 2 and Fig. 1).
Fig. 1.
Results of a univariate analysis of risk factors for EMD in patients with AL. A-D: Comparison of age, WBC, BM blast percentage, and PB blast percentage between two groups of AL patients, with and without EMD. E-H: Comparison of 18F-FDG PET/CT semi-quantitative parameters, including FL SUVmax, FL SUVmean, FL SUVpeak, and the SUVmax FL/liver ratio, between the two groups of AL patients, with and without EMD. EMD extramedullary disease, WBC white blood cells, BM bone marrow, PB peripheral blood, FL focal lesion, SUV standardized uptake value. *, P < 0.05; **, P < 0.01
Subsequently, a multivariate analysis was performed to identify independent factors that influenced the presence of EMD in patients with AL. The analysis revealed that age at the time of initial diagnosis of AL (odds ratio [OR]: 0.888, P < 0.01) and the percentage of blasts of PB (OR: 1.061, P < 0.05) were significant independent risk factors for the appearance of EMD (Table 2).
Table 2.
Multivariate analysis of factors that affect patients with AL with EMD
| Variable | B | Standard Error | OR | OR (95% CI) | P | |
|---|---|---|---|---|---|---|
| lower | upper | |||||
| Age | −0.119 | 0.043 | 0.888 | 0.815 | 0.966 | 0.006** |
| WBC | 0.001 | 0.021 | 1.001 | 0.961 | 1.042 | 0.965 |
| BM blasts/% | −0.042 | 0.033 | 0.959 | 0.898 | 1.023 | 0.202 |
| PB blasts/% | 0.059 | 0.027 | 1.061 | 1.006 | 1.119 | 0.03* |
| FL SUVmax | −0.409 | 0.907 | 0.664 | 0.112 | 3.925 | 0.652 |
| FL SUVmean | 0.936 | 1.195 | 2.551 | 0.245 | 26.534 | 0.433 |
| FL SUVpeak | 0.766 | 1.283 | 2.152 | 0.174 | 26.622 | 0.55 |
| SUVmax FL/liver ratio | −0.745 | 0.859 | 0.475 | 0.088 | 2.559 | 0.386 |
EMD extramedullary disease, WBC white blood cells, BM bone marrow, PB peripheral blood, FL focal lesion, SUV standardized uptake value, OR odds ratio, CI confidence interval, B regression coefficient. *, P < 0.05; **, P < 0.01
Analysis of the efficacy of clinical parameter determine whether patients with AL have EMD
It is evident from the analysis that both age and the percentage of PB blasts are significant independent risk factors for the development of EMD in patients with AL. In this context, ‘B’ represents the regression coefficient, indicating the extent of influence each predictor has on the occurrence of EMD. The formula to calculate the values of the joint predictor (denoted as L) is as follows.
 |
ROC curves for age, percentage of PB blast, and combined predictors were plotted to assess the probability of an extramedullary lesion. Preliminary analysis of the ROC curve suggests that the combined predictor showed relatively high accuracy in determining the presence of EMD, with an AUC of 0.82. According to the calculations, the predictor achieved a specificity of 88.2% but a sensitivity of only 65.2% (Fig. 2).
Fig. 2.
ROC curve of age, percentage of PB blasts, and their combination to predict the presence of EMD in patients with AL. PB peripheral blood, CI confidence interval, AUC area under the curve
Subgroup analysis of patients with AL with EMD (EMD+)
Statistics of extramedullary infiltration sites in patients with AL with EMD
In our study, we identified 25 patients with AL with EMD, involving a total of 47 organs infiltrated by extramedullary lesions. These organs were classified according to the location of the infiltration. The sites most frequently affected were the lymph nodes (LN) (42.55%), followed by the lungs (10.64%), the anterior mediastinum (8.51%), the nasopharynx and oral cavity (8.51%) and the skin (6.38%). The intestines, liver and kidneys each had two case of infiltration, accounting for 4.26% of the total. The spleen, pancreas, tonsils, parotid gland and uterus each had a single case of infiltration, accounting for 2.13% of the total (Fig. 3).
Fig. 3.
Statistical map of extramedullary infiltration sites in the EMD (+) subgroup
Independent predictors of OS in patients with AL presenting EMD
In our study, LN were the most frequent site of extramedullary lesions (Fig. 3). Consequently, we classified patients into subgroups based on the presence or absence of additional organ infiltration beyond LN and performed Kaplan–Meier survival analyses. Within the EMD(+) subgroup, patients with infiltration of organs beyond LN (n = 16) exhibited a significantly lower OS compared to those without such additional infiltration (n = 9), with median survival times of 157 days and 806 days, respectively (P < 0.05) (Fig. 4B).
Fig. 4.
Kaplan–Meier survival analysis of OS in patients with AL. Kaplan–Meier survival analysis of OS in EMD(−) and EMD(+) groups (A). Kaplan–Meier survival analysis OS in EMD(+) subgroup (B). EMD extramedullary disease, LN lymph nodes
Subsequent univariate Cox regression analysis of the semi-quantitative parameters of PET/CT and clinical utility in the EMD(+) subgroup revealed that several metrics were significantly correlated with OS. These included the SUVmax BM/liver ratio (Hazard ratio [HR], 1.932; 95% CI, 1.012–3.687; P < 0.05), EMD SUVmax (HR, 1.121; 95% CI, 1.009–1.246; P < 0.05), EMD SUVmean (HR, 1.226; 95% CI, 1.002–1.501; P < 0.05), and percentage of PB blasts (HR, 0.975; 95% CI, 0.954–0.997; P < 0.05) (Table 3). A subsequent multivariate analysis identified the SUVmax BM/liver ratio (HR, 2.372; 95% CI, 1.079–5.214) as an independent risk factor that affects OS in patients with AL with EMD (P < 0.05). However, infiltration of other organs in addition to LN, EMD SUVmax, EMD SUVmean, and the percentage of PB blast did not emerge as independent risk factors for OS in these patients (Fig. 5).
Table 3.
Univariate analysis of factors affecting OS in patients with AL with EMD
| Variable | B | Standard Error | HR | HR (95.0%CI) | P | |
|---|---|---|---|---|---|---|
| lower | upper | |||||
| WBC | −0.038 | 0.027 | 0.963 | 0.913 | 1.015 | 0.157 |
| BM blasts/% | −0.001 | 0.009 | 0.999 | 0.981 | 1.017 | 0.906 |
| PB blasts/% | −0.025 | 0.011 | 0.975 | 0.954 | 0.997 | 0.024* |
| β2M | < 0.001 | < 0.001 | 1 | 1 | 1 | 0.254 |
| LDH | < 0.001 | < 0.001 | 1 | 0.999 | 1 | 0.234 |
| BM SUVmax | 0.059 | 0.09 | 1.061 | 0.889 | 1.265 | 0.512 |
| BM SUVmean | 0.125 | 0.14 | 1.133 | 0.861 | 1.491 | 0.373 |
| BM SUVpeak | 0.09 | 0.116 | 1.094 | 0.871 | 1.375 | 0.438 |
| SUVmax BM/liver ratio | 0.659 | 0.33 | 1.932 | 1.012 | 3.687 | 0.046* |
| FL SUVmax | 0.145 | 0.079 | 1.156 | 0.99 | 1.349 | 0.066 |
| FL SUVmean | 0.259 | 0.142 | 1.296 | 0.981 | 1.713 | 0.068 |
| FL SUVpeak | 0.128 | 0.11 | 1.137 | 0.916 | 1.411 | 0.245 |
| EMD SUVmax | 0.114 | 0.054 | 1.121 | 1.009 | 1.246 | 0.034* |
| EMD SUVmean | 0.204 | 0.103 | 1.226 | 1.002 | 1.501 | 0.048* |
| EMD SUVpeak | 0.129 | 0.069 | 1.138 | 0.994 | 1.302 | 0.061 |
| MTV of EMD | 0.002 | 0.001 | 1.002 | 1 | 1.005 | 0.075 |
| TLG of EMD | < 0.001 | < 0.001 | 1 | 1 | 1.001 | 0.071 |
EMD extramedullary disease, SUV standardized uptake value, WBC white blood cells, BM bone marrow, PB peripheral blood, β2M Beta-2-microglobulin, LDH lactate dehydrogenase, FL focal lesion, B regression coefficient, OS overall survival, HR hazard ratio, CI confidence interval. *, P < 0.05
Fig. 5.
Multivariate analysis of factors affecting OS in patients with AL with EMD. OS overall survival, CI confidence interval, SUV standardized uptake value, BM bone marrow, PB peripheral blood, LN lymph nodes, EMD extramedullary disease
Correlation between PET/CT parameters and clinical and laboratory characteristics
Correlations between 18F-FDG PET/CT parameters (SUVmax, SUVmean, SUVpeak of FL and BM) and clinical characteristics (BM blast percentage, PB blast percentage, WBC, β2M and LDH were evaluated using Spearman’s rank correlation analysis (Table 4). The results indicated that BM blast percentage was significantly associated with several semi-quantitative PET/CT parameters, including BM SUVmax (r = 0.382, P = 0.013), BM SUVmean (r = 0.407, P = 0.007), BM SUVpeak (r = 0.437, P = 0.004), the SUVmax BM/liver ratio (r = 0.311, P = 0.045), FL SUVmax (r = 0.453, P = 0.003), FL SUVmean (r = 0.452, P = 0.003), and the FL SUVpeak (r = 0.473, P = 0.002). Additionally, β2M was significantly associated with BM SUVmean (r = 0.318, P = 0.048) and the BM SUVpeak (r = 0.354, P = 0.027).Furthermore, a significant correlations were found between LDH and several semi-quantitative PET/CT parameters such as BM SUVmax (r = 0.324, P = 0.034), BM SUVmean (r = 0.309, P = 0.043), BM SUVpeak (r = 0.306, P = 0.046), FL SUVmax (r = 0.323, P = 0.035) and the FL SUVpeak (r = 0.333, P = 0.029).
Table 4.
Correlation between PET/CT parameters and clinical and laboratory characteristics
| Variable | BM blasts/% | PB blasts/% | WBC | β2M | LDH | |
|---|---|---|---|---|---|---|
| BM SUVmax | r | 0.382 | −0.107 | −0.066 | 0.281 | 0.324 |
| P | 0.013* | 0.51 | 0.672 | 0.083 | 0.034* | |
| BM SUVmean | r | 0.407 | −0.122 | −0.088 | 0.318 | 0.309 |
| P | 0.007** | 0.453 | 0.571 | 0.048* | 0.043* | |
| BM SUVpeak | r | 0.437 | −0.073 | −0.051 | 0.354 | 0.306 |
| P | 0.004** | 0.653 | 0.742 | 0.027* | 0.046* | |
| SUVmax BM/liver ratio | r | 0.311 | −0.137 | −0.072 | 0.058 | 0.26 |
| P | 0.045* | 0.399 | 0.644 | 0.726 | 0.093 | |
| FL SUVmax | r | 0.453 | −0.026 | −0.128 | 0.159 | 0.323 |
| P | 0.003** | 0.873 | 0.408 | 0.333 | 0.035* | |
| FL SUVmean | r | 0.452 | −0.043 | −0.14 | 0.149 | 0.296 |
| P | 0.003** | 0.793 | 0.366 | 0.366 | 0.054 | |
| FL SUVpeak | r | 0.473 | −0.011 | −0.107 | 0.222 | 0.333 |
| P | 0.002** | 0.945 | 0.491 | 0.173 | 0.029* |
PET/CT positron emission tomography/computed tomography, BM bone marrow, FL focal lesion, PB peripheral blood, SUV standardized uptake value, WBC white blood cell, β2M Beta-2-microglobulin, LDH lactate dehydrogenase. *, P < 0.05; **, P < 0.01
Discussion
In our study, extramedullary AL lesions were identified at a variety of sites and some patients exhibited multiorgan infiltration. The most frequently infiltrated sites were LN at 42.55% and lungs at 10.64%. Our analysis revealed significant differences between groups with and without EMD in factors such as age at diagnosis, WBC counts, BM blast percentage, PB blast percentage, FL SUVmax, FL SUVmean, FL SUVpeak, and SUVmax FL/liver ratio (P < 0.05). Age at diagnosis and percentage of PB blasts were identified as independent risk factors for the development of EMD. Subsequently, the combined predictor ROC curve, calculated and analyzed from these independent risk factors, demonstrated high diagnostic precision to identify extramedullary lesions, with an AUC of 0.82, a specificity of 88.2%, but a sensitivity of only 65.2%. In the subgroup of patients with EMD, the SUVmax BM/liver ratio was an independent factor that affected OS. Furthermore, our correlation analysis indicated that certain laboratory parameters were associated with PET/CT parameters.
Extramedullary manifestations of AL are complex, diverse, and relatively common. A study by Stölzel et al. reported a prevalence of 22% of patients with AML combined with EMD and a total of 65 EMD manifestations. Notably, this prevalence was higher when detected with PET/CT scans than previously reported or hypothesized [5]. In our study, 39 of 44 newly diagnosed patients with AL presented with typical symptoms of hematopoietic suppression, such as blood abnormalities, malaise, and dizziness, while only five exhibited extramedullary infiltration as an initial symptom (Table 1). However, subsequent 18F-FDG PET/CT scans revealed that 25 patients had EMD, with 16 experiencing organ infiltration beyond LN. This discrepancy indicates that clinical symptoms in patients with AL do not consistently align with the actual extent of leukemic infiltration, highlighting a significant mismatch between observed symptoms and the true tumor burden. In cases of AL where clinically observable and significant extramedullary lesions are present but are considered isolated, the lack of whole-body scanning for systemic involvement leads to a significant underestimate of the extent of leukemia involvement [14]. Currently, clinicians may overlook the evaluation of potential EMD throughout the body at the initial diagnosis of patients with AL. Even when independent EMD is detected, clinicians often rely on localized CT or MRI for assessment, whereas PET/CT is not as extensively utilized in leukemia as it is in lymphomas and MM. At the onset of the disease, attention to extramedullary lesions may not be prioritized, and these conditions are easily overlooked in clinical practice. Patients are often diagnosed only when they present with unbearable clinical symptoms, at which point extramedullary lesions are typically large or widespread and the prognosis at this stage is generally poor [15]. Conventional imaging modalities such as CT and MRI are extensively used in the clinical setting for detecting EMD due to their cost-effectiveness and convenience. However, their limited sensitivity and scanning range often result in the inability to detect small occult lesions. Studies have shown that the incidence of patients with AML with EMD identified by 18F-FDG PET/CT scans is more than double that found in clinical examinations, with detection rates of 65% versus 31% [4]. This suggests that timely systemic evaluation using PET/CT is essential at the initial diagnosis or at the time of lesion detection. Such evaluations can partially compensate for the oversight of potential extramedullary lesions in clinical evaluations. Consequently, it is important and necessary to improve the use of PET/CT in the management of AL.
Chang et al. discovered that in AML, all abnormal cytogenetic groups, except the 11q23 abnormality, were not associated with the development of extramedullary infiltrates [16]. Our result did not show any significant differences between patients with AL with EMD and those without, in terms of karyotypes or the presence of high-risk gene mutations (Table 1). For patients with EMD(+), WBC counts, the percentage of BM blasts and the percentage of PB blasts were significantly higher than in patients with AL without EMD (Fig. 1B-D). β2M and LDH levels are known to reflect tumor burden in patients [17, 18]. In our study, we found that β2M and LDH levels were correlated with several semi-quantitative PET/CT parameters. Specifically, LDH was associated with FL SUVmax and FL SUVpeak, both of which were significantly elevated in patients with EMD (Fig. 1E, G). These findings suggest that patients with a higher tumor burden at initial diagnosis are more likely to develop EMD. This correlation was further supported by the fact that the percentage of PB blasts (OR: 1.061, P < 0.05) was an independent risk factor for the development of EMD in patients with AL, as established in a multifactorial follow-up study (Table 2). In addition to PB blasts, multivariate analysis identified age at diagnosis (OR:0.888, P < 0.05) as another independent factor that influences the development of EMD. This suggests that younger patients at the initial diagnosis of AL with a higher tumor load are more likely to show EMD early in the course of the disease. Further analysis of the combined effects of age and PB blast counts was conducted to determine the presence of EMD. Combined predictors demonstrated a specificity of 88.2% to detect EMD in patients with AL, although the sensitivity was limited to 65.2%. This highlights the challenge of predicting 18F-FDG PET/CT results solely on the basis of laboratory results, primarily due to low sensitivity.
In our study, the CR rate in patients with EMD was lower than the overall CR rate, at 44% compared to 54.5% (Table 1). However, no significant differences in OS were observed between patients with EMD(+) and EMD(−), with median OS values of 384 days and 226 days, respectively (P = 0.291, Fig. 4A). This result aligns with findings from Ganzel et al. [12]. Subsequent univariate and multivariate Cox regression analyses of clinical utility and the semi-quantitative parameters of PET/CT did not identify any independent risk factors significantly associated with OS (Supplementary Table 3).Further subgroup analysis of patients with EMD(+) indicated that those with organ infiltration beyond LN experienced a significantly worse prognosis, with a median OS of 157 days compared to 806 days for those without such infiltration (Fig. 4B). However, a subsequent multifactorial analysis revealed that this factor was not an independent risk factor for patients in the EMD(+) subgroup. However, this finding suggests that patients initially diagnosed with AL, particularly those with organ infiltration beyond LN, should be approached with caution in clinical settings due to their potential association with a poorer prognosis. Multivariate analysis of factors affecting OS in the EMD(+) subgroup identified the SUVmax BM/liver ratio as an independent risk factor affecting OS, with an HR of 2.372 and a 95% CI of 1.079–5.214 (Fig. 5). Similar to the semi-quantitative parameter of our study, A. Paschali et al. demonstrated that the SUVmax pelvis/liver ratio was significantly lower in the group of good differentiation in patients with MM [19]. Our study highlights the prognostic value of 18F-FDG PET/CT in patients with EMD, particularly demonstrating that a higher SUVmax BM/liver ratio is associated with shorter OS and poorer prognosis. These findings emphasize the need for further prospective studies to confirm the prognostic significance of this imaging parameter. Additionally, the results underscore the potential role of 18F-FDG PET/CT in improving the clinical management of acute AL at the time of initial diagnosis. By providing detailed information on metabolic activity and identifying extramedullary lesions, PET/CT enables a more comprehensive assessment, which can significantly enhance prognostic evaluation and guide treatment strategies for patients with AL.
Although AL is a systemic malignancy, the distribution of leukemia cells can be heterogeneous and localized. Consequently, BM biopsy results may not offer a complete picture of the patient’s systemic tumor burden [20, 21]. In this context, 18F-FDG PET/CT offers a significant advantage as a noninvasive method for the comprehensive evaluation of AL. Although we emphasize the importance of using 18F-FDG PET/CT in AL, current treatment guidelines do not recommend different therapeutic approaches for cases of EMD at initial diagnosis [22, 23]. We believe that an early PET/CT scan could help clinicians assess the extent and location of extramedullary lesions or potential lesions, thus preventing delays in addressing symptoms when patients later present with extramedullary manifestations. This could be crucial for the timely detection of tumor progression and the adjustment of treatment plans, highlighting the value of 18F-FDG PET/CT as a non-invasive diagnostic tool. Furthermore, detection and treatment of lesions early, rather than after significant progression, is more manageable, especially as advances in targeted radiation therapy and local ablation have improved outcomes in leukemia [2].
Our study was a single-center retrospective study with inherent bias in sample selection and featured a small sample size. Additionally, AL was not categorized into specific subtypes or associated with particular treatment options for further analysis, which may have introduced bias into the results. However, we applied strict inclusion criteria and excluded patients with irregular treatment to minimize bias in this aspect. Future studies should involve larger cohorts and the design of prospective clinical trials to more accurately determine the clinical relevance of 18F-FDG PET/CT in AL. Furthermore, while PET/CT is a valuable diagnostic tool, its high cost may impose a financial burden on patients, representing a significant limitation in its routine use for AL.
In summary, the SUVmax BM/liver ratio is an independent risk factor for OS in patients with EMD, underscoring the prognostic value of 18F-FDG PET/CT in AL. 18F-FDG PET/CT serves as a non-invasive and highly sensitive diagnostic tool that significantly enhances our ability to perform a comprehensive assessment of the burden of systemic tumors at the initial diagnosis of AL. Consequently, 18F-FDG PET/CT enables clinicians to comprehensively and accurately assess the initial disease status of patients, facilitating the development of more personalized treatment strategies and improving patient outcomes in AL.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
The contributions of participants in our center are sincerely acknowledged by the authors.
Author contributions
All authors contributed to the study’s conception and design. J.M.F. performed the data analyses and wrote the manuscript. J.C. performed an image examination of patients. X.Q.L. P.P.L. collected and analyzed the patient data. F.L.Z., Y.H. and X.Y.L. contributed to the conception of the study and reviewed the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding
This work was supported by the Natural Science Foundation of China (NSFC) program [grant number 81900116, 82370176, 82000127, 82200254], and the Zhongnan Hospital of Wuhan University discipline construction plat form project [grant number 202021, PDJH202217].
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all individual participants included in the study.
Patient consent for publication
Study participants provided their consent for the publication of any data and associated images and all identifying patient data was removed.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Xiaoyan Liu, Email: liuxiaoyan@znhospital.cn.
Yong He, Email: vincentheyong@163.com.
Fuling Zhou, Email: zhoufuling@whu.edu.cn.
References
- 1.H. Sung, J. Ferlay, R.L. Siegel, M. Laversanne, I. Soerjomataram, A. Jemal, F. Bray, Global Cancer Statistics 2020, GLOBOCAN estimates of incidence and Mortality Worldwide for 36 cancers in 185 countries. Cancer J. Clin. 71, 209–249 (2021). 10.3322/caac.21660 [DOI] [PubMed] [Google Scholar]
- 2.C.D. Dinardo, H.P. Erba, S.D. Freeman, A.H. Wei, Acute myeloid leukaemia. Lancet (London England). 401, 2073–2086 (2023). 10.1016/S0140-6736(23)00108-3 [DOI] [PubMed] [Google Scholar]
- 3.M. Onciu, Acute lymphoblastic leukemia. Hematol. Oncol. Clin. N. Am. 23, 655–674 (2009). 10.1016/j.hoc.2009.04.009 [DOI] [PubMed] [Google Scholar]
- 4.A.-S.W.I. Cribe, M. Steenhof, C.W. Marcher, H. Petersen, H. Frederiksen, L.S. Friis, Extramedullary disease in patients with acute myeloid leukemia assessed by 18F-FDG PET. Eur. J. Haematol. 90, 273–278 (2013). 10.1111/ejh.12085 [DOI] [PubMed] [Google Scholar]
- 5.F. Stölzel, T. Lüer, S. Löck, S. Parmentier, F. Kuithan, M. Kramer, N.S. Alakel, K. Sockel, F. Taube, J.M. Middeke, J. Schetelig, C. Röllig, T. Paulus, J. Kotzerke, G. Ehninger, M. Bornhäuser, M. Schaich, K. Zoephel, The prevalence of extramedullary acute myeloid leukemia detected by 18FDG-PET/CT: final results from the prospective PETAML trial. Haematologica. 105, 1552–1558 (2020). 10.3324/haematol.2019.223032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.J. Hillengass, S. Usmani, S.V. Rajkumar, B.G.M. Durie, M.-V. Mateos, S. Lonial, C. Joao, K.C. Anderson, R. García-Sanz, E. Riva, J. Du, N. Van De Donk, J.G. Berdeja, E. Terpos, E. Zamagni, R.A. Kyle, J. San Miguel, H. Goldschmidt, S. Giralt, S. Kumar, N. Raje, H. Ludwig, E. Ocio, R. Schots, H. Einsele, F. Schjesvold, W.-M. Chen, N. Abildgaard, B.C. Lipe, D. Dytfeld, B.M. Wirk, M. Drake, M. Cavo, J.J. Lahuerta, S. Lentzsch, International myeloma working group consensus recommendations on imaging in monoclonal plasma cell disorders. Lancet Oncol. 20, e302–e312 (2019). 10.1016/S1470-2045(19)30309-2 [DOI] [PubMed] [Google Scholar]
- 7.J.V. Alvarez, G.K. Belka, T.-C. Pan, C.-C. Chen, E. Blankemeyer, A. Alavi, J.S. Karp, L.A. Chodosh, Oncogene pathway activation in mammary tumors dictates FDG-PET uptake. Cancer Res. 74, 7583–7598 (2014). 10.1158/0008-5472.CAN-14-1235 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.P. Seam, M.E. Juweid, B.D. Cheson, The role of FDG-PET scans in patients with lymphoma. Blood. 110, 3507–3516 (2007). 10.1182/blood-2007-06-097238 [DOI] [PubMed] [Google Scholar]
- 9.E. Zamagni, C. Nanni, F. Patriarca, E. Englaro, P. Castellucci, O. Geatti, P. Tosi, P. Tacchetti, D. Cangini, G. Perrone, M. Ceccolini, A. Brioli, S. Buttignol, R. Fanin, E. Salizzoni, M. Baccarani, S. Fanti, M. Cavo, A prospective comparison of 18F-fluorodeoxyglucose positron emission tomography-computed tomography, magnetic resonance imaging and whole-body planar radiographs in the assessment of bone disease in newly diagnosed multiple myeloma. Haematologica. 92, 50–55 (2007). 10.3324/haematol.10554 [DOI] [PubMed] [Google Scholar]
- 10.H.J. Cho, S.-H. Jung, J.-C. Jo, Y.J. Lee, S.E. Yoon, S.-S. Park, D.Y. Kim, H.-J. Shin, Y.-C. Mun, J.H. Yi, H.J. Kim, D.J. Kim, H.S. Lee, S.H. Bae, C.M. Hong, S.Y. Jeong, J.-J. Min, S.K. Sohn, C.-K. Min, K. Kim, J.-J. Lee, J.H. Moon, K.M.M.W. Party, Development of a new risk stratification system for patients with newly diagnosed multiple myeloma using R-ISS and 18F-FDG PET/CT. Blood Cancer J. 11, 190 (2021). 10.1038/s41408-021-00577-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.M.A. Aljama, M.H. Sidiqi, F.K. Buadi, M.Q. Lacy, M.A. Gertz, A. Dispenzieri, D. Dingli, E. Muchtar, A.L. Fonder, S.R. Hayman, M.A. Hobbs, W.I. Gonsalves, R.M. Warsame, T. Kourelis, Y.L. Hwa, P. Kapoor, R.A. Kyle, N. Leung, R.S. Go, S.V. Rajkumar, S.K. Kumar, Utility and prognostic value of 18 F-FDG positron emission tomography-computed tomography scans in patients with newly diagnosed multiple myeloma. Am. J. Hematol. 93, 1518–1523 (2018). 10.1002/ajh.25279 [DOI] [PubMed] [Google Scholar]
- 12.C. Ganzel, J. Manola, D. Douer, J.M. Rowe, H.F. Fernandez, E.M. Paietta, M.R. Litzow, J.-W. Lee, S.M. Luger, H.M. Lazarus, L.D. Cripe, P.H. Wiernik, M.S. Tallman, Extramedullary Disease in Adult Acute myeloid leukemia is common but lacks independent significance: analysis of patients in ECOG-ACRIN Cancer Research Group trials, 1980–2008. J. Clin. Oncology: Official J. Am. Soc. Clin. Oncol. 34, 3544–3553 (2016). 10.1200/JCO.2016.67.5892 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.J.C. Byrd, R.B. Weiss, D.C. Arthur, D. Lawrence, M.R. Baer, F. Davey, E.S. Trikha, A.J. Carroll, R. Tantravahi, M. Qumsiyeh, S.R. Patil, J.O. Moore, R.J. Mayer, C.A. Schiffer, C.D. Bloomfield, Extramedullary leukemia adversely affects hematologic complete remission rate and overall survival in patients with t(8;21)(q22;q22): results from Cancer and Leukemia Group B 8461. J. Clin. Oncology: Official J. Am. Soc. Clin. Oncol. 15, 466–475 (1997). 10.1200/JCO.1997.15.2.466 [DOI] [PubMed] [Google Scholar]
- 14.I. Cunningham, B. Kohno, 18 FDG-PET/CT: 21st century approach to leukemic tumors in 124 cases. Am. J. Hematol. 91, 379–384 (2016). 10.1002/ajh.24287 [DOI] [PubMed] [Google Scholar]
- 15.Z. Zhao, Y. Hu, J. Li, Y. Zhou, B. Zhang, S. Deng, Applications of PET in diagnosis and prognosis of Leukemia. Technol. Cancer Res. Treat. 19, 1–12 (2020). 10.1177/1533033820956993 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.H. Chang, J. Brandwein, Q.-L. Yi, K. Chun, B. Patterson, B. Brien, Extramedullary infiltrates of AML are associated with CD56 expression, 11q23 abnormalities and inferior clinical outcome. Leuk. Res. 28, 1007–1011 (2004). 10.1016/j.leukres.2004.01.006 [DOI] [PubMed] [Google Scholar]
- 17.M.L. Sorror, B.E. Storer, A.T. Fathi, A.T. Gerds, B.C. Medeiros, P. Shami, A.M. Brunner, M.A. Sekeres, S. Mukherjee, E. Peña, M. Elsawy, S. Wardyn, J. Whitten, R. Moore, P.S. Becker, J.S. Mccune, F.R. Appelbaum, E.H. Estey, Development and validation of a Novel Acute myeloid leukemia-composite model to Estimate risks of Mortality. JAMA Oncol. 3, 1675–1682 (2017). 10.1001/jamaoncol.2017.2714 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.W. Wulaningsih, L. Holmberg, H. Garmo, H. Malmstrom, M. Lambe, N. Hammar, G. Walldius, I. Jungner, T. Ng, M.V. Hemelrijck, Serum lactate dehydrogenase and survival following cancer diagnosis. Br. J. Cancer. 113, 1389–1396 (2015). 10.1038/bjc.2015.361 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.A. Paschali, E. Panagiotidis, T. Triantafyllou, V. Palaska, K. Tsirou, E. Verrou, E. Υiannaki, D. Markala, A. Papanikolaou, A. Pouli, P. Konstantinidou, V. Chatzipavlidou, E. Terpos, E. Katodritou, A proposed index of diffuse bone marrow [18F]-FDG uptake and PET skeletal patterns correlate with myeloma prognostic markers, plasma cell morphology, and response to therapy. Eur. J. Nucl. Med. Mol. Imaging. 48, 1487–1497 (2021). 10.1007/s00259-020-05078-1 [DOI] [PubMed] [Google Scholar]
- 20.T. Endo, N. Sato, K. Koizumi, M. Nishio, K. Fujimoto, T. Sakai, K. Kumano, M. Obara, K. Minauchi, T. Koike, Localized relapse in bone marrow of extremities after allogeneic stem cell transplantation for acute lymphoblastic leukemia. Am. J. Hematol. 76, 279–282 (2004). 10.1002/ajh.20106 [DOI] [PubMed] [Google Scholar]
- 21.T. Maeda, S. Kosugi, H. Ujiie, K. Osumi, T. Fukui, H. Yoshida, H. Kashiwagi, J. Ishikawa, Y. Tomiyama, Y. Matsuzawa, Localized relapse in bone marrow in a posttransplantation patient with t(6;9) acute myeloid leukemia. Int. J. Hematol. 77, 522–525 (2003). 10.1007/BF02986623 [DOI] [PubMed] [Google Scholar]
- 22.H. Döhner, A.H. Wei, F.R. Appelbaum, C. Craddock, C.D. Dinardo, H. Dombret, B.L. Ebert, P. Fenaux, L.A. Godley, R.P. Hasserjian, R.A. Larson, R.L. Levine, Y. Miyazaki, D. Niederwieser, G. Ossenkoppele, C. Röllig, J. Sierra, E.M. Stein, M.S. Tallman, H.-F. Tien, J. Wang, A. Wierzbowska, B. Löwenberg, Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN. Blood. 140, 1345–1377 (2022). 10.1182/blood.2022016867 [DOI] [PubMed] [Google Scholar]
- 23.N. Gökbuget, N. Boissel, S. Chiaretti, H. Dombret, M. Doubek, A. Fielding, R. Foà, S. Giebel, D. Hoelzer, M. Hunault, D.I. Marks, G. Martinelli, O. Ottmann, A. Rijneveld, P. Rousselot, J. Ribera, R. Bassan, Management of ALL in Adults: 2023 ELN Recommendations from a European Expert Panel. Blood, 1903–1930 (2024). 10.1182/blood.2023023568 [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.