Table 2 Univariate Cox regression analysis for PFS in all subjects.

From: Cluster analysis of autoencoder-extracted FDG PET/CT features identifies multiple myeloma patients with poor prognosis

Variable

Categories

Hazard ratio

95% CI

P value

Log-rank test

Sex

Female

   

0.113

Male

1.365

0.927–2.011

0.115

Age

 < 58

   

0.006

58–67

1.334

0.829–2.147

0.235

67 ≤ 

2.126

1.319–3.429

0.002

Chemotherapy regimen

VTD

   

 < 0.001

VMP

2.472

1.544–3.958

 < 0.001

Other

1.650

1.021–2.668

0.041

Radiotherapy

No

   

0.624

Yes

0.844

0.426–1.671

0.626

Autologous stem cell transplantation

No

   

 < 0.001

Yes

0.426

0.289–0.626

 < 0.001

R-ISS stage

I

   

0.004

II

1.250

0.758–2.062

0.382

III

2.375

1.345–4.193

0.003

Hypercalcemia

No

   

0.712

Yes

1.111

0.633–1.951

0.714

Renal insufficiency

No

   

0.090

Yes

1.457

0.947–2.243

0.087

Anemia

No

   

0.008

Yes

1.684

1.143–2.482

0.008

Bone lesions

No

   

0.990

Yes

1.003

0.581–1.733

0.990

Extramedullary disease

No

   

0.545

Yes

0.852

0.506–1.432

0.544

Unsupervised cluster

A

   

 < 0.001

B

0.931

0.612–1.416

0.739

C

2.694

1.556–4.662

 < 0.001

Supervised cluster

A

   

0.002

B

0.958

0.612–1.485

0.848

C

2.234

1.359–3.676

0.001

MTV

Low

   

 < 0.001

Moderate

0.660

0.389–1.120

0.124

High

1.865

1.212–2.869

0.005

  1. PFS progression-free survival, VMP bortezomib, melphalan, and prednisone, R-ISS, revised multiple myeloma international staging system, MTV metabolic tumor volume, CI confidence interval, VTD bortezomib, thalidomide, dexamethasone.
Back to article page