RU2821774C1 - Method for prediction of overall survival of patients with chronic lymphocytic leukemia stage a-c in dynamics of disease - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention relates to oncohematology and can be used for prediction of overall survival rate of patients with chronic lymphocytic leukaemia A-C stage according to Binet in five-year dynamics of disease. Entering the formula L = XY, where X is the GFR parameter of the patient by formula CKD-EPI, Y = 76 during the first year of the disease, Y = 75 during the second year of the disease, Y = 70 during the third year of the disease, Y = 68 during the fourth and fifth years of the disease. At L ≥ 0 patient will overcome median survival rate according to Binet classification. If L < 0, the patient will not overcome the median survival rate according to the Binet classification.

EFFECT: method provides the possibility of increasing the accuracy of prediction of the overall survival rate of the patients with chronic lymphocytic leukaemia by calculating the glomerular filtration rate in the patient with chronic lymphocytic leukaemia by formula CKD-EPI from 1st to 5^th year of the clinical course of the disease.

1 cl, 2 tbl, 5 ex

Description

The invention relates to medicine, in particular to oncohematology, and can be used to predict the overall survival of patients with chronic lymphocytic leukemia in the dynamics of the disease.

Chronic lymphocytic leukemia (CLL) is today one of the most common nosologies from the group of lymphoproliferative diseases [1]. Currently, the generally accepted staging systems for CLL are K. Rai (1975) [2], J. Binet (1981) [3], which allow one to calculate the median patient survival depending on the size of the tumor mass. However, these systems do not reflect the prognosis of the disease over time and do not take into account the comorbid pathology that is observed in elderly patients at whom CLL is diagnosed. In real clinical practice, a patient's overall survival (OS) may differ significantly from the calculated median survival. For this reason, the reliability of the results of predicting OS based on the volume of the tumor mass remains insufficiently high. In view of this, identifying the reasons that influence the OS rate of patients with CLL is of particular relevance.

As a prototype of the present invention, we used a method for predicting the overall survival of male patients with chronic lymphocytic leukemia in stages A-C, where the formula was used: L=X-76.5, where X is the patient’s GFR parameter at the time of diagnosis, calculated using the MDRD formula, while in In case of L≥0, the patient will overcome the median survival according to the Binet classification; if L<0, the patient will not overcome the median survival according to the Binet classification [4].

Despite the sufficient information content, the above method has a number of disadvantages: it uses an outdated formula for calculating GFR MDRD [5], does not apply to females, has a recognition accuracy of only 85% and is limited to the prognosis at the stage of diagnosis, not reflecting the dynamics of the disease. The proposed method for predicting the OS of patients with CLL is free of these disadvantages: it uses the GFR parameter calculated using the CKD-EPI formula, is applicable to both sexes, provides higher class separability, and is applicable within 5 years from diagnosis.

The problem to be solved by the claimed invention is to create a method for predicting deviations in the overall survival of patients with CLL in stages A-C from the calculated parameters of the median survival according to the Binet classification.

This problem is solved by calculating the glomerular filtration rate in a patient with CLL using the CKD-EPI formula from years 1 to 5 of the disease.

The implementation of the proposed method is carried out as follows. In patients with stage A-C CLL according to Binet, peripheral venous blood is collected and creatinine levels are measured using standard biochemical methods. The patient's glomerular filtration rate (GFR) is calculated using the CKD-EPI formula [6]:

GFR=141*min (serum creatinine/kappa, 1)alpha*max

(Serum creatinine/kappa, 1)1.209*0.993 Age*Gender*Race

for women the following meanings are used:

gender=1.018; alpha=-0.329; kappa=0.7;

for men the following values are used: gender=1; alpha=-0.411; kappa=0.9;

for representatives of the Negroid race: coefficient “race” = 1.159.

GFR calculator: https://www.kidney.org/professionals/KDOQI/gfr_calculator

The proposed method was developed based on the results of a study of 207 patients with Binet stage A-C CLL with known OS, observed in the hematology department of the Ulyanovsk Regional Clinical Hospital. During each year of follow-up, the maximum value of the patient's GFR was taken into account.

The sample subjects were divided into two non-overlapping classes K1 (actual survival less than the predicted median OS of 85 patients) and K2 (actual survival greater than or equal to the predicted median OS of 122 patients).

The ordered sequence of values for this quantitative characteristic was divided into two non-overlapping intervals [s ₁ ; с ₂ ], (с ₂ ; с ₃ ], according to the value of criterion [7]

Where - the number of attribute values of objects from the class K _i , i=1,2, respectively, in the intervals [with ₁ ; с ₂ ] and (с ₂ ; c ₃ ]. The expression in the left brackets (1) represents intra-class similarity, in the right - inter-class difference. The set of acceptable values of the criterion (0; 1]. If each interval contains all the values of the attribute of objects of the same class , then the criterion is equal to 1.

The most significant separability between classes was in terms of GFR according to CKD-EPI with a value of criterion (1) equal to 0.8068, so this indicator was subsequently studied in more detail.

The value G=(c ₂ +b)/2 was used as the boundary (threshold) between classes, where b(b>c ₂ ) is the value of the attribute closest to c ₂ from (c ₂ ;c ₃ ].

The values of the boundaries of the characteristic intervals at the maximum value of criterion (1) are presented in Table 1. The number of objects decreases over time, because Patients who died were automatically excluded from consideration.

As can be seen from Table 1, the values of GFR according to CKD-EPI when predicting the overall survival of patients with CLL monotonically decreases over time from 75.5 to 66.

To justify the choice of threshold between classes, the stability of dividing the feature into non-overlapping intervals was calculated [8].

As can be seen from Table 2, the stability value of dividing a characteristic into non-overlapping intervals decreases monotonically.

Prediction of overall survival of patients with CLL in stage A-C according to Binet was made using the formula L=X-Y, where X is the patient’s GFR parameter according to the CKD-EPI formula, Y=76 during the first year of the disease, Y=75 during the second year of the disease, Y =70 during the third year of the disease, Y=68 during the fourth and fifth years of the disease. If L≥0, the patient will overcome the median survival rate according to the Binet classification; if L<0, the patient will not overcome the median survival rate according to the Binet classification.

For illustration, we provide clinical examples.

Clinical example 1. Patient T., aged 55 years, during a routine medical examination in March 2002, a general blood test revealed absolute lymphocytosis, and was referred for consultation to a hematologist. He did not actively complain. Upon examination, the general condition is satisfactory. The sizes of the peripheral lymph nodes, liver and spleen are normal. In the general blood test: red blood cells - 4.81×10 ¹² /l, hemoglobin - 148 g/l, leukocytes - 21.1×10 ⁹ /l, lymphocytes - 63% (absolute number - 13.3×10 ⁹ /l ), platelets - 211×10 ⁹ /l. According to the myelogram, bone marrow infiltration with lymphoid elements was 48%. Based on clinical and laboratory data, a diagnosis of chronic lymphocytic leukemia, stage A according to Binet, was made.

Dynamics of GFR during the patient's observation and his survival.

Clinical example 2. Patient K. At the age of 68 years in May 2015, a diagnosis was made: chronic lymphocytic leukemia, stage B according to Binet. At the time of diagnosis, there were complaints of weakness and increased fatigue, periodically causeless low-grade fever, enlargement of the submandibular and axillary lymph nodes. During an objective examination, these groups of lymph nodes were enlarged to 2.5-3 cm, soft-elastic consistency, hepatosplenomegaly was recorded (liver +2 cm, spleen +5 cm). In the general blood test: red blood cells - 5.2 × 10 ¹² / l, hemoglobin - 120 g / l, leukocytes - 34.2 × 10 ⁹ / l, mature lymphocytes - 68% (absolute number - 23.3 × 10 ⁹ / l), platelets - 194 × 10 ⁹ / l. When studying the myelogram, bone marrow infiltration with mature lymphocytes was 63%.

Dynamics of GFR during the patient's observation and his survival.

Clinical example 3. Patient 3. In September 2015, at the age of 69 years, a diagnosis was made: chronic lymphocytic leukemia, stage B according to Binet. At the time of diagnosis, the patient complained of weakness, increased fatigue, periodic rises in body temperature to subfebrile levels, and enlarged submandibular lymph nodes. On examination: general condition is satisfactory. Peripheral lymph nodes, accessible to palpation, are enlarged to 2 cm, soft-elastic consistency, liver +3 cm, spleen +6 cm. In the general blood test: red blood cells - 4.72 × 10 ¹² / l, hemoglobin - 136 g/l, leukocytes -24.8×10 ⁹ /l, mature lymphocytes - 73% (absolute number - 18.1×10 ⁹ /l), platelets - 182×10 ⁹ /l. When studying the myelogram, bone marrow infiltration with mature lymphocytes was 59%.

Dynamics of GFR during the patient's observation and his survival.

Clinical example 4. Patient N. In May 2016, at the age of 75 years, a diagnosis was made: chronic lymphocytic leukemia, stage C according to Binet. At the time of diagnosis, the patient complained of severe weakness, fatigue, periodic rises in body temperature to febrile levels, an increase in the volume of the abdomen, axillary, supraclavicular, submandibular lymph nodes. On examination: the general condition is of moderate severity. Peripheral lymph nodes are enlarged to 3 cm, soft-elastic consistency, liver +6 cm, spleen +16 cm. In the general blood test: red blood cells - 2.72 × 10 ¹² / l, hemoglobin - 106 g/l, leukocytes - 100, 8×10 ⁹ /l, mature lymphocytes - 93% (absolute number - 93.7×10 ⁹ /l), platelets - 98×10 ⁹ /l. In a myelogram study, bone marrow infiltration with mature lymphocytes was 89%.

Dynamics of GFR during patient observation and survival.

Clinical example 5. Patient B. In May 2017, at the age of 70 years, a diagnosis was made: chronic lymphocytic leukemia, stage C according to Binet. At the time of diagnosis, she complained of weakness, fatigue, frequent occurrences of infectious diseases of the upper respiratory tract with a rise in body temperature to subfebrile levels (more than 4 times a year), an increase in the volume of the abdomen, axillary, supraclavicular, submandibular, and posterior cervical lymph nodes. On examination: the general condition is of moderate severity. Peripheral lymph nodes are enlarged to 3-4 cm, soft-elastic consistency, liver +5 cm, spleen +12 cm. In the general blood test: red blood cells - 2.82 × 10 ¹² /l, hemoglobin - 105 g/l, leukocytes - 92.4×10 ⁹ /l, mature lymphocytes - 83% (absolute number - 76.7×10 ⁹ /l), platelets - 105×10 ⁹ /l. When studying the myelogram, bone marrow infiltration with mature lymphocytes was 74%.

Dynamics of GFR during patient observation and survival.

Advantages of the proposed method:

1. Uses the GFR parameter calculated using the CKD-EPI formula;

2. Applicable to both sexes;

3. Gives high separability of classes;

4. Apply within 5 years from the diagnosis of CLL.

Thus, the proposed method makes it possible to more accurately predict the overall survival of patients with CLL stage A-C according to Binet, which affects the optimization of therapy and the quality of life of patients.

Used Books

1. Diagnostic algorithms and treatment protocols for diseases of the blood system. Ed. V.G. Savchenko. In two volumes. Volume 2. - M.: Praktika, 2018. - 1264 p.

2. Rai KR, Sawitsky A, Cronkite EP et al. Clinical staging of chronic lymphocytic leukemia. Blood 1975; 46:219-234.

3. Binet J. L., Auquier A, Dighiero G et al. A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis // Cancer. - 1981. - T. 48, issue. 1. - pp. 198-206.

4. Patent RU 2725877 C1. A method for predicting the overall survival of male patients with chronic lymphocytic leukemia in stages A-C. / Markovtseva M.V., Zguralskaya E.N. Bull. No. Publ. 7.7.2020

5. Levey AS, Stevens LA. Estimating GFR using the CKD Epidemiology Collaboration (CKD-EPI) creatinine equation: more accurate GFR estimates, lower CKD prevalence estimates, and better risk predictions. Am J Kidney Dis 2010;55(4):622-7. doi:10.1053/j.ajkd.2010.02.337

6. Clinical recommendations. Chronic kidney disease (CKD). Nephrology.2021;25(5): 10-82.

7. Zguralskaya E.N. Stability of partitioning data into intervals in recognition problems and searching for hidden patterns // Izvestia Samarsk. scientific center of Russia acad. Sci. - 2018. - T. 20, No. 4(3). - pp. 451-455.

8. Ignatiev N.A. Calculation of generalized indicators and intelligent data analysis // Automation and telemechanics. - 2011. - No. 5. - pp. 183-190.

Claims (1)

A method for predicting the overall survival of patients with chronic lymphocytic leukemia stage A-C according to Binet in the five-year dynamics of the disease, where the formula L=X-Y is introduced, where X is the patient’s GFR parameter according to the CKD-EPI formula, Y=76 during the first year of the disease, Y=75 in during the second year of the disease, Y=70 during the third year of the disease, Y=68 during the fourth and fifth years of the disease; with L≥0 the patient will overcome the median survival according to the Binet classification; with L<0 the patient will not overcome the median survival according to the Binet classification.