Sora Kang and Hyungwoo Cho contributed equally to this work.
Prognostic Index for Natural Killer Lymphoma (PINK) is the most widely accepted prognostic model for patients with extranodal natural killer/T-cell lymphoma (ENKTL) treated with non-anthracycline–based therapy. We aimed to evaluate the prognostic implications of serum β-2 microglobulin (β2M) in the context of PINK and proposed a new prognostic model.
A total of 138 patients who were newly diagnosed with ENKTL and treated with non-anthracycline-based chemotherapy were identified. The cut-off value of high serum β2M was calculated by maximal-chi square methods (4.1 mg/L). A new prognostic model incorporating serum β2M into PINK was proposed and validated in an independent validation cohort (n=88).
The patients’ median age was 53.5 years (range, 19 to 80 years). Patients with high serum β2M levels had significantly worse overall survival (OS) and progression-free survival (PFS). In multivariate analysis, high serum β2M was an independent adverse prognostic factor for OS. A new PINK-B (Prognostic Index for Natural Killer Lymphoma-serum β-2 microglobulin) model stratified patients into three groups with distinct OS and PFS in the training cohort (3-year OS, 84.1% [95% confidence interval, 75.1 to 94.2], 46.8% [36.1 to 60.8] and 17.6% [6.3 to 49.2] for the low-, intermediate, and high-risk groups, respectively; 3-year PFS, 70.6% [59.4 to 83.8], 35.9% [25.9 to 49.8], and 7.35% [1.1 to 46.7] for the low-, intermediate-, and high-risk groups, respectively). The PINK-B model was further validated in an independent cohort.
Serum β2M is an independent prognostic factor for ENKTL patients. The new serum β2M-based prognostic model may be useful for identifying ultra-high-risk patients, and it can easily be adopted into daily clinical practice.
Extranodal natural killer/T-cell lymphoma (ENKTL) is a rare subtype of mature T/natural killer (NK) cell lymphoma characterized by a strong association with Epstein-Barr virus (EBV) infection and a predominant involvement pattern of the nasal cavity and nasopharynx [
The introduction of non-anthracycline–based chemotherapy including L-asparaginase has markedly improved the survival outcome of ENKTL patients, and it is currently used as standard frontline therapy [
β-2 microglobulin (β2M) is a low molecular weight protein that is a component of the major histocompatibility complex class I antigen, and it can be seen on most nucleated cells [
Between February 2003 and June 2018, 211 patients who were newly diagnosed with ENKTL were identified in the database of the Asan lymphoma Registry, Asan Medical Center, Seoul, South Korea. The diagnosis was confirmed by an experienced pathologist according to the World Health Organization classification [
A radioimmunoassay kit (Immunotech, Inc., Prague, Czech Republic) was used to measure serum β2M in both the training and validation cohort (reference range, 1.0 to 2.4 mg/L). β2M was measured as follows. Serum samples were collected in tubes with EDTA and separated from cells by centrifugation within 3 hours after collection. Then, the serum sample, 50 μL of the calibrator and 500 μL of tracer (148 kBq 125I-labeled β2M in buffer with bovine serum albumin, sodium azide [< 0.1%], and a dye) were sequentially added to anti-β2M monoclonal antibody-coated tubes and mixed. Then, incubation was carried out for 90 minutes at 18–25°C with shaking (> 280 rpm). After incubation, the supernatants were discarded, and the bound radioactivity of the precipitates was determined using a gamma counter. A standard curve was constructed and the serum β2M levels were measured from the curve by interpolation [
PINK is based on four risk factors: age greater than 60, Ann Arbor stage III or IV, involvement of distant LN, and non-nasal type. Low-, intermediate-, and high-risk groups were classified according to the sum of the number of risk factors (0, low risk; 1, intermediate risk; 2/3/4, high risk) [
A chi-square test or Fisher exact test was used to analyze the correlation between categorical variables. OS was defined as the time from the date of diagnosis to death from any cause. Progression-free survival (PFS) was defined as the time from the diagnosis to the date of disease progression or death, whichever occurred first. OS and PFS were estimated by the Kaplan-Meier method, and the log-rank test was used to compare the OS or PFS between the risk groups. To determine the optimal cut-off value for serum β2M to most accurately predict the OS outcomes, the maximal chi-square method was used [
The baseline characteristics of the patients in the training cohort (n=138) are described in
The cut-off value of serum β2M was determined as 4.1 mg/L using the maximal chi-square method, and the baseline characteristics of the patients with high (≥ 4.1 mg/L, n=26) and low (< 4.1 mg/L, n=112) serum β2M were compared. The proportions of patients with ECOG PS 2–4 (23.1% of 26, p=0.001), stage III or IV (61.5% of 26, p=0.001), distant LN involvement (65.4% of 26, p < 0.001), bone marrow involvement (26.9% of 26, p=0.006), and elevated serum LDH (92.3%, p < 0.001) were significantly higher in the high-serum β2M group compared to the low-serum β2M group. In the low serum β2M group, more patients received concurrent chemo-radiotherapy (n=75) compared with the high-serum β2M group (n=7) (p < 0.001).
With a median follow-up duration of 8.0 years among the surviving patients (range, 2.5 to 16.4), the median OS and PFS in entire patients were 11.95 years (95% confidence interval [CI], 4.17 to not available [NA]) and 2.32 years (95% CI, 1.57 to 7.41), respectively (
Considering the results that the high serum β2M level further stratified patients with a worse prognosis among the patients in the high-risk group of PINK, we re-classified patients in the PINK high-risk group based on their serum β2M levels. Among the PINK high-risk group (n=44), 27 (61.3% of 44) patients were redefined as the high risk with low serum β2M group, and 17 patients (38.6% of 44) were re-classified as the high risk with high serum β2M group. The median OS for the low-, intermediate-, and high-risk with low serum β2M-, and high-risk with high serum β2M groups were not reached (95% CI, 11.95 to NA), 7.48 years (95% CI, 1.77 to NA), 2.19 years (95% CI, 1.13 to NA), and 0.38 years (95% CI, 0.23 to 2.16), respectively (
Based on these results, we proposed a new prognostic model incorporating serum β2M into PINK, PINK-B. Patients who were originally classified as PINK low- and intermediate-risk groups were classified as PINK-B low- and intermediate groups, respectively. Among the patients in the high-risk group of PINK, patients with low serum β2M were redefined as the PINK-B intermediate-risk group, and patients with high serum β2M were redefined as the PINK-B high-risk group. In the training cohort, PINK-B discriminated the patients into three risk groups, which had distinct OS and PFS (p < 0.001 for both). The 3-year OS according to PINK-B was 84.14% (95% CI, 75.14 to 94.21), 46.88% (95% CI, 36.11 to 60.84) and 17.65 (95% CI, 6.32 to 49.28) in the low, intermediate, and high-risk groups, respectively (
The prognostic value of PINK-B was further validated in the independent validation cohort (n=88). The patient characteristics of the validation cohort are described in
Several previous studies have reported the prognostic value of serum β2M in patients with ENKTL [
Previously, elevated serum β2M has been widely evaluated as a prognostic factor in numerous types of lymphoproliferative disorders [
The exact mechanisms underlying the association of a poor prognosis of patients with ENKTL and high serum β2M is still unclear. Our results showed that the patients with high serum β2M were related to high tumor burdens, such as disseminated bone disease, advanced stage, and elevated LDH. These findings are in line with previous reports that a high serum β2M is a reflection of a high tumor burden [
The PINK and PINK–Epstein-Barr virus (PINK-E) are the most widely accepted prognostic models for ENKTL patients treated with non-anthracycline–based chemotherapy. Several studies have suggested that the PINK-E model may be more useful than the PINK model to stratify high-risk patients, probably due to the strong association between EBV DNA positivity and the prognosis of ENKTL [
There are several limitations of the present study. As anticipated for any retrospective study, this study is susceptible to selection bias. In addition, the cutoff value of serum β2M in this study was relatively high compared with previous studies [
In conclusion, serum β2M is an independent prognostic factor for patients with ENKTL. Our new prognostic model incorporating serum β2M into PINK, PINK-B, may be useful for identifying ultra-high-risk patients, and it can be easily adopted into daily clinical practice. This model should be validated in additional large-scale independent cohorts.
Supplementary materials are available at Cancer Research and Treatment website (
This study was reviewed and approved by the Institutional Review Board (IRB approval number: 2019-1476) of the Asan Medical Center (Seoul, South Korea) and Samsung Medical Center (Seoul, South Korea). The requirement of informed consent was waived by the IRBs because of the study’s retrospective nature.
Conceived and designed the analysis: Suh C.
Collected the data: Kang S, Lee K, Kang EH, Park JS, Lee YS, Park CS, Go H, Huh J, Ryu JS, Lee SW, Kim SJ, Kim WS, Yoon SE, Ko YH, Suh C.
Contributed data or analysis tools: Kang S, Cho H, Lee K, Kang EH, Park JS, Lee YS, Park CS, Go H, Huh J, Ryu JS, Lee SW, Kim SJ, Kim WS, Yoon SE, Ko YH, Suh C.
Performed the analysis: Kang S, Cho H, Suh C.
Wrote the paper: Kang S, Cho H, Suh C.
Review the manuscript: Lee K, Kang EH, Park JS, Lee YS, Park CS, Go H, Ryu JS, Lee SW, Kim SJ, Kim WS, Yoon SE, Ko YH.
Conflict of interest relevant to this article was not reported.
Overall survival (A) and progression-free survival (B) for all patients. Overall survival (C) and progression-free survival (D) according to serum β-2 microglobulin (β2M) 4.1 mg/L.
Overall survival (A) and progression-free survival (B) according to Prognostic Index for Natural Killer Lymphoma.
Prognostic value of serum β2M in the risk groups determined by PINK. Overall survival according to serum β2M levels (< 4.1 mg/L vs. ≥ 4.1 mg/L) in the PINK low-risk group (A), PINK intermediate group (B), and PINK high-risk group (C). β2M, β-2 microglobulin; PINK, Prognostic Index for Natural Killer Lymphoma.
(A) Overall survival of patients according to PINK and incorporating serum β2M for the PINK high-risk groups. Overall survival (B) and progression-free survival (C) according to PINK-B. β2M, β-2 microglobulin; PINK, Prognostic Index for Natural Killer Lymphoma; PINK-B, Prognostic Index for Natural Killer Lymphoma–serum β-2 microglobulin.
Overall survival (A) and progression-free survival (B) according to PINK-B in the validation cohort. PINK-B, Prognostic Index for Natural Killer Lymphoma–serum β-2 microglobulin.
Baseline characteristics of the included patients
Total (n=138) | Low β2M (< 4.1) (n=112) | High β2M (≥ 4.1) (n=26) | p-value | |
---|---|---|---|---|
|
||||
≤ 60 | 94 (68.1) | 77 (68.8) | 17 (65.4) | 0.740 |
> 60 | 44 (31.9) | 35 (31.2) | 9 (34.6) | |
Median (range) | 53.5 (19–80) | 53.5 (19–80) | 54 (25–79) | 0.682 |
|
||||
Male | 86 (62.3) | 71 (63.4) | 15 (57.7) | 0.589 |
Female | 52 (37.7) | 41 (36.6) | 11 (42.3) | |
|
||||
0 or 1 | 129 (93.5) | 109 (97.3) | 20 (76.9) | 0.001 |
≥ 2 | 9 (6.5) | 3 (2.7) | 6 (23.1) | |
|
||||
I or II | 93 (67.4) | 83 (74.1) | 10 (38.5) | 0.001 |
III or IV | 45 (32.6) | 29 (25.9) | 16 (61.5) | |
|
||||
None/Regional | 99 (71.7) | 90 (80.4) | 9 (34.6) | < 0.001 |
Distant | 39 (28.3) | 22 (19.6) | 17 (65.4) | |
|
||||
No | 114 (82.6) | 96 (85.7) | 18 (69.2) | 0.080 |
Yes | 24 (17.4) | 16 (14.3) | 8 (30.8) | |
|
||||
No | 123 (89.8) | 104 (93.7) | 19 (73.1) | 0.006 |
Yes | 14 (10.2) | 7 (6.3) | 7 (26.9) | |
|
||||
Normal | 78 (56.9) | 76 (68.5) | 2 (7.7) | < 0.001 |
Elevated | 59 (43.1) | 35 (31.5) | 24 (92.3) | |
|
||||
Normal (< 1.5 mg/dL) | 137 (99.2) | 112 (100) | 25 (96.1) | 0.188 |
Elevated (≥ 1.5 mg/dL) | 1 (0.7) | 0 | 1 (3.8) | |
|
||||
Detectable | 71 (51.4) | 49 (43.7) | 22 (84.6) | < 0.001 |
Non-detectable | 61 (44.2) | 58 (51.7) | 3 (11.5) | |
Unknown | 6 (4.3) | 5 (4.5) | 1 (3.8) | |
|
||||
Chemotherapy alone | 56 (40.6) | 37 (33.0) | 19 (73.1) | < 0.001 |
CCRT alone | 11 (7.97) | 10 (8.9) | 1 (3.8) | |
CCRT followed by chemotherapy | 71 (51.4) | 65 (58.0) | 6 (23.1) | |
|
||||
PINK low | 57 (41.3) | 52 (46.4) | 5 (19.2) | < 0.001 |
PINK intermediate | 37 (26.8) | 33 (29.5) | 4 (15.4) | |
PINK high | 44 (31.9) | 27 (24.1) | 17 (65.4) | |
|
||||
0 or 1 | 92 (66.6) | 82 (74.5) | 10 (35.7) | < 0.001 |
≥ 2 | 46 (33.3) | 28 (25.5) | 18 (64.3) | |
|
||||
SMILE | 35 (25.3) | 22 (19.6) | 13 (50.0) | 0.012 |
VIDL | 67 (48.5) | 59 (52.7) | 8 (30.8) | |
VIPD | 15 (10.8) | 13 (11.6) | 2 (7.7) | |
MIDLE | 4 (2.89) | 4 (3.6) | 0 | |
IMEP | 3 (2.1) | 3 (2.7) | 0 | |
DeVIC | 1 (0.72) | 1 (0.9) | 0 | |
DHAP | 1 (0.72) | 0 | 1 (3.8) | |
ESHAP | 1 (0.72) | 0 | 1 (3.8) |
Values ae presented as number (%) or median (range). β2M, serum β-2 microglobulin; CCRT, concurrent chemoradiotherapy; ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase; PINK, Prognostic Index for Natural Killer Lymphoma; PS, performance status.
Chemotherapy regimen abbreviations: SMILE, corticosteroid, methotrexate, ifosfamide, L-asparaginase, and etoposide; VIDL, etoposide, ifosfamide, dexamethasone, and L-asparaginase; VIPD, etoposide, ifosfamide, cisplatin, and dexamethasone; MIDLE, methotrexate, ifosfamide, dexamethasone, L-asparaginase, and etoposide; IMEP, ifosfamide, methotrexate, etoposide, and prednisolone; DeVIC, dexamethasone, etoposide, ifosfamide, and carboplatin; DHAP, dexamethasone, cisplatin, and cytarabine; ESAHP, etoposide, corticosteroid, cytarabine, and cisplatin.
Factors associated with overall survival and progression-free survival
Variable | Overall survival | Progression-free survival | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|||||||||||
Univariate analysis | Multivariate analysis | Univariate analysis | Multivariate analysis | |||||||||
|
|
|
|
|||||||||
HR | 95% CI | p-value | HR | 95% CI | p-value | HR | 95% CI | p-value | HR | 95% CI | p-value | |
Female | 1.03 | 0.62–1.70 | 0.912 | - | - | - | 0.87 | 0.55–1.37 | 0.541 | - | - | - |
|
||||||||||||
B symptoms presence | 1.68 | 1.02–2.77 | 0.042 | - | - | - | 1.54 | 0.97–2.43 | 0.065 | - | - | - |
|
||||||||||||
ECOG PS 2–4 | 7.55 | 3.53–16.14 | < 0.001 | - | - | - | 5.71 | 2.71–12.03 | < 0.001 | - | - | - |
|
||||||||||||
Age ≥ 60 yr | 1.77 | 1.08–2.90 | 0.024 | 1.87 | 1.10–3.18 | 0.019 | 1.52 | 0.96–2.41 | 0.073 | 1.49 | 0.93–2.41 | 0.092 |
|
||||||||||||
Stage III–IV | 4.38 | 2.67–7.19 | < 0.001 | 5.69 | 2.88–11.22 | < 0.001 | 3.37 | 2.15–5.28 | < 0.001 | 2.93 | 1.83–4.70 | < 0.001 |
|
||||||||||||
Non-nasal type | 3.11 | 1.80–5.39 | < 0.001 | 3.13 | 1.56–6.28 | 0.001 | 2.69 | 1.61–4.49 | < 0.001 | 2.14 | 1.26–3.65 | 0.004 |
|
||||||||||||
Distant LN involvement | 3.22 | 1.96–5.28 | < 0.001 | - | - | - | 2.97 | 1.88–4.68 | < 0.001 | - | - | - |
|
||||||||||||
Serum LDH ≥ 250 U/L | 2.72 | 1.80–4.08 | < 0.001 | - | - | - | 2.35 | 1.74–4.81 | < 0.001 | - | - | - |
|
||||||||||||
Serum β2M ≥ 4.1 mg/L | 3.31 | 1.94–5.65 | < 0.001 | 2.87 | 1.58–5.21 | < 0.001 | 2.89 | 1.74–4.81 | < 0.001 | 2.16 | 1.28–3.66 | 0.003 |
B2M, β-2 microglobulin; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; LDH, lactate dehydrogenase; LN, lymph node; PS, performance status.
Stratified models for overall survival and progression-free survival
3-Year overall survival | 3-Year progression-free survival | HR (95% CI) | AIC | C-index (95% CI) | |
---|---|---|---|---|---|
|
|||||
Low | 84.1 (75.1–94.2) | 70.6 (59.4–83.8) | Reference | 562.6 | 0.80 (0.73–0.87) |
Intermediate | 54.0 (40.1–72.7) | 43.2 (29.9–62.5) | 3.12 (1.52–6.39) | ||
High | 29.5 (18.7–46.6) | 19.3 (10.5–35.8) | 6.85 (3.50–13.31) | ||
|
|||||
Low | 84.1 (75.1–94.2) | 70.6 (59.4–83.8) | Reference | 556.9 | 0.82 (0.75–0.89) |
Intermediate | 46.8 (36.1–60.8) | 35.9 (25.9–49.8) | 3.72 (1.93–7.15) | ||
High | 17.6 (6.32–49.2) | 7.35 (1.16–46.7) | 12.6 (5.9–27.04) |
AIC, Akaike Information Criteria; CI, confidence interval; HR, hazard ratio; PINK, Prognostic Index for Natural Killer Lymphoma; PINK-B, Prognostic Index for Natural Killer Lymphoma serum β-2 microglobulin.