Tóm tắt Luận án Evaluate the value of uterine artery Doppler ultrasound and prognostic factors to predict methotrexate resistance in low - Risk gestational trophoblastic neoplasia

are resistant to MTX early and need to be treated with EMA-CO At the present time, nearly all GTN patients are cured with chemotherapy, but incorporating a UAPI into the FIGO scoring system will help in selecting earlier and more accurate patients who need combined chemotherapy. The main purpose is to increase the effectiveness of treatment, reduce the total duration of treatment, reduce psychological stress, reduce toxicity of chemicals and soon return fertility to GTN patients who still want to give birth. Many authors such as Agarwal, Long, Sita Lumsden said that it would be helpful to supplement the UAPI with the FIGO risk scoring system. It will be more objective and valuable when the UAPI is studied in a multicenter, if it is proven to be an effective and valuable exploration method at GTN treatment centers. To date, in Vietnam, there has not been any study to use Doppler ultrasound to predict the MTX resistance of low-risk GTN patients. Therefore, we conducted this study to evaluate the effectiveness of Doppler ultrasound combined with related factors to predict the resistance to MTX and choose the appropriate regimen to treat the disease. CHAPTER 2: RESEARCH SUBJECTS AND METHODS 2.1. Time and site of research At the NHOG from 01/2015 to 09/2017. 2.2. Research subjects The patient was diagnosed as a low-risk GTN according to the FIGO 2002 classification, preserving uterus and treated with MTX. 7 2.2.1. Selection criteria Patients participating in the study must satisfy the following conditions: - GTN patients with a score of 0 to 6, were classified as low-risk according to FIGO prognostic score in 2002. - Having diagnosis of GTN post-molar: a history of molar is diagnosed and treated at the NHOG or other hospital and one of the following criteria: + GTN may be diagnosed when the plateau of hCGlasts for 4 measurements over a period of 3 weeks or longer, that is days 1,7,14,21. + GTN may be diagnosed when there is a rise of hCG on three consecutive weekly measurements, over a period of two weeks or longer, days 1,7,14. + GTN is diagnosed if there is histologic diagnosis of choriocarcinoma. + GTN is diagnosed when the hCG level remains elevated for 6 months or more. - GTN after miscarriage, abortion, postpartum and ectopic pregnancy. - Patients with liver, kidney and hematological function within normal limits. - Patients who voluntarily participate in the study and are treated at the NHOG until they are discharged or finished treatment because they are no longer capable of specific treatment. - The patient agrees to use contraception methods during treatment. 2.2.2. Exclusion criteria Patients who have any one among the following issues will be excluded from the study: - Not a GTN patient. - GTN but did not receive treatment at the hospital, gave up treatment or circumstances did not allow treatment and monitored until the end of the course. - GTN needs radiation therapy. - Having a history of other cancers being treated. - Allergy to MTX. - Not following the treatment regimen. - Incomplete or missing ultrasound results. 8 Table 2.1. Prognostic Scoring System 2002 by FIGO for GTN Prognostic factors Scores 0 1 2 4 Age (years) < 40 ≥ 40 Antecedent pregnancy mole abortion term Interval months from index pregnancy (month) 12 Pretreatment serum hCG < 10 3 10 3 - 10 4 10 4 - 10 5 > 10 5 Largest tumour size (including uterus) 3 - 4 ≥ 5 Site of metastases lung spleen, kidney Gastro- intestinal liver, brain Number of metastases 1 - 4 5 - 8 > 8 Previous failed chemotherapy single drug ≥ 2 drugs 2.3. Research methods 2.3.1. Research design: Descriptive prospective study 2.3.2. Research sample size: is calculated using the following formula 2 2 )2/α1( .. d qpZ n = n: number of patients p: resistance rate to MTX chemotherapy (following the study of Phan Chí Thành (2012) resistance rate is 24,5%), so p = 0,245. q = 1- p. d: relative bias, chosen d = 0,06 α: Statistical significance level, chosen α = 0,05 Z from z table at α = 0,05, so taken Z = 1,96. Thus, the number of research subjects is at least 198. We selected 204 low-risk GTN patients. 2.3.3. Research facilities 2.3.3.1. Drugs used in research 9 Drugs used in research: Methotrexat 50mg / 5ml of Ebewe, licensed VN3-63-15 according to No. 413 of the Drug Administration of 2015. 2.3.3.2. Research ultrasound machine GE's Voluson 730 ultrasound is used to conduct research data collection and UA Doppler measurement. The device is equipped with a 3.5 MHz abdominal ultrasound probe. It has pulse Doppler, Color Doppler and Power enhanced Doppler. 2.3.3.3. Person performing ultrasound research Doppler ultrasound is performed by a diagnostic imaging doctor or obstetrician- gynecologist who has a gynecological ultrasound certificate. Doctors conducting ultrasound research are well-informed and know the procedure of abdominal gynecological ultrasound and bilateral uterine artery Doppler ultrasound. 2.4. Research steps 2.4.1. The process of enrolling patients in the study 2.4.1.1. Identify GTN patients - Collect patients with GTN diagnosed who are treated at the GIO meeting the selection criteria. - GTN patients who are admitted to hospital are assessed prognosis according to FIGO 2002. 2.4.1.2. Full physical examination 2.4.1.3. Antecedent question 2.4.1.4. Gynecological examination 2.4.1.5. Paraclinical tests - Chest radiography to detect lung metastases. - General abdominal ultrasound to detect metastases in the liver and kidney. - Brain CT or MRI scan to detect brain metastases when clinically suggestive. - Hematological test: + Blood count test: Number of red blood cells, quantification of Hemoglobin. + The number of white blood cells, neutrophils. + Blood biochemistry: assessment of liver function and kidney function. 2.4.1.6. Conseling patients participating in the study - Patients who satisfy the selection criteria, desire and able to participate in the study will sign a consent form to participate in research. 2.4.2. Steps to conduct research Step 1: Pelvic ultrasound and UAPI Doppler measurement once for patients before starting MTX treatment. - The patient have ultrasound done when bladder half full to measure uterine volume, uterine myometrial invasion size , then UAPI Doppler ultrasound on both sides. - UAPI index retained for analysis is a lower index (reflecting maximum variation compared to normal values). - The original Doppler ultrasound results will be retained for inclusion in the study. 10 - The patient was then used alternately MTX / FA regimen. Step 2: Implement the chemotherapy monotherapy regimen - GTN patients will be given 50mgMTX, deep intramuscular injection on 1,3,5,7 days with 5mg FA intramuscular injection on 2,4,6,8 days. - Repeat regimen every 14 days. - Monitor the response to MTX by quantifying βhCG concentration before each course of MTX/FA. - Before each course of treatment, patients were assessed red blood cells, white blood cells, platelets, liver, kidney function, and undesirable effects. - Monitor response to MTX chemotherapy. - Patients who are resistant to MTX will be swiched to treat with combined chemotherapy regimen EMACO, EMAEP. - Patients are treated until βhCG <5 IU/l. - After negative βhCG, patient were sent to outpatient monitoring (<5 IU / l) 3 consecutive times, 1 week apart, it will be considered as a cured. 2.5. Method of evaluating results 2.5.1. Criteria of MTX cured After treatment with MTX monotherapy, the patient achieves: - Serum βhCG <5 I /L serum for 3 consecutive weeks. - Complete assessed when serum βhCG levels returned to normal (<5 IU/l) + Normal liver and kidney function + Normal blood formula + There is no sign of myelosuppression or leukopenia + No new metastase appear + No more metastase lesions in the lung on chest X-ray film + The myometrial invasion at the uterus often disappears. In some cases, sclerosis may disappear after a few months. + Normal abdominal ultrasound 2.5.2. Criteria of chemotherapy resistance After each cycle of chemotherapy treatment (MTX, EMACO, EMAEP), the patient presents with: - βhCG increases, does not decrease or decrease less than 10% after 2 weeks. - New metastatic need to be changed to chemotherapy regimen. - Patients on a regimen due to chemical side effects (mouth ulcers or allergies) are not considered chemotherapy resistant. 2.5.3. Follow up after treatment After the end of treatment, serum βhCG levels are monitored weekly for the first 4 weeks, then every 2 weeks for up to 3 months, followed by monthly for the first year, then every 3 months on second year and every 6 months to 1 year until the end of life. 2.6. Ethics in research - The research is for scientific purposes only, and is approved by the Ethics council of biomedical research, NHOG. - The participation of women in this research is completely voluntary. - All information about participants or information from medical records is kept confidential. 11 2.7. Collect, enter and process data The data are processed on computers by the method of medical statistics under the Stata program. CHAPTER 3: RESEARCH RESULTS 3.1. Characteristics and treatment results of patients of LR-GTN Place of residence: patients living in rural and mountainous areas account for a high proportion (62.8%). Number of living children: most patients have no children or only 1 child (84.3%). History of abortion: The proportion of patients without a history of abortion accounts for a high proportion (56.4%). 3.1.1. Treatment results of patients of LR-GTN - The rate of response to MTX chemotherapy is quite high, accounting for 72.55%. - 27.45% of patients were resistant to MTX and had to switch regimens. 3.1.2. Treatments after MTX resistance Table 3.1. Treatments after MTX resistance Methods (n = 56) Cycles (min-max) Duration(day) (min-max) Rate (%) Hysterect omy + chemo MTX (N=2) 2 ± 0 (2 - 2) 83,5 ± 13,4 (74 - 93) 3,6 EMACO (N=31) 4,0 ± 2,3 (1 - 9) 151,8 ± 76,7 (66 - 428) 55,4 Combined + chemo EMACO (N=18) 2,9 ± 1,8 (1 - 4) 124,2 ± 46,8 (56 - 213) 28,5 EMACO+ EMAEP (N=5) 3,4 ± 2,3 (1 - 7) 3,6 ± 1,8 (1 - 6) 208,0 ± 63,9 (128 - 265) 12,5 - The group of patients who received surgery and EMACO treatment accounted for the majority. 3.1.3. Results of treatment of patients of LR-GTN - The total cured rate of LR-GTN patients was 99.5%. - 1 patient was not cured in the study 3.2. Characteristics of prognostic factors in GTN patients - Age: on average, 26.3 years old (16-39 years old). - Index pregnancy: Majority is molar 81.4%. 12 - Latent time: Most are <4 months, the average is 2.1 months. - History of MTX treatment: Only 3 cases GTN relapse, accounting for 1.5%. - Distribution of metastase location and stage: Most have no metastases and in stage 1 of FIGO, accounting for 95.6%. - Number of metastase: a few patients have ≥ 1 metastase, 4.4%. - Largest tumor size: The majority of patients without tumors, accounting for 66.7%. Among 68 patients with tumors (including uterine, lung metastatic and vaginal metastases), the most common group of tumor size was 3-4 cm, accounting for 17.2%. The largest size of an average tumor is 3.4 ± 1.1 cm (1.3 - 6.9 cm). 3.2.1. βhCG pretreatment of GTN patients Bảng 3.2. Nồng độ βhCG trước điều trị của bệnh nhân UNBN NCT Nồng độ βhCG trước điều trị (IU/l) n % < 1.000 87 42,7 1.000 - < 10.000 64 31,3 10.000 - < 100.000 49 24,0 ≥ 100.000 4 2,0 Tổng 204 100,0 - The majority of patients have pre-treatment hCG levels <1000 IU / l. 3.3.2. Distribution of FIGO score of GTN patients Figure 3.2. Distribution of FIGO score of GTN patients - Distribution of FIGO score of GTN patients irregular, focused on groups with low FIGO scores of 0 - 3 points. 3.3. Characteristics of uterine artery Doppler of GTN patients 3.4.1. Features of bilateral doppler ultrasound at uterin artery Table 3.2. Features of bilateral doppler ultrasound at uterin artery Value Doppler Higher PI lower PI p1 p2 24,5 28,9 20,6 16,2 6,4 3,4 0 10 20 30 0 1 2 3 4 ≥5 FIGO score P er ce n ta g e (% ) 13 X ± SD Median X ± SD Median Mean Median PSV (cm/s) 64,1 ± 30,6 57,5 61,9 ± 30,0 54,6 0,46 0,037 EDV (cm/s) 18,2 ± 22,3 9,25 12,1 ± 17,5 6,1 0,002 < 0,001 PI 2,14 ± 1,19 2,02 2,82 ± 1,47 2,63 0,001 < 0,001 RI 0,78 ± 0,18 0,82 0,85 ± 0,16 0,89 0,001 < 0,001 S/D 5,42 ± 5,71 4,11 6,83 ± 7,31 5,13 0,07 < 0,001 Median PSV of the lower PI side is higher than the high PI side, statistically significant, p< 0.05. The difference between the median value of PI, RI and S/D between the two parties is statistically significant, p <0.05. 3.4.2. Features of doppler ultrasound at uterin artery of GTN patients Table 3.3. Features of Doppler ultrasound at uterine artery Value Doppler X ± SD Median min max PSV (cm/s) 63,0 ± 27,9 55,65 24,3 229,5 EDV (cm/s) 15,1 ± 19,1 7,48 00,0 121,5 PI 2,5 ± 1,3 2,32 0,54 6,21 RI 0,8 ± 0,2 0,85 0,38 1,00 S/D 5,1 ± 3,4 4,27 1,71 26,84 - Median systolic velocity at at uterin artery is 55,65 cm/s. - Median diastolic velocity at uterin artery is 7,48 cm/s. 3.4. Evaluation of factors related to prognosis of MTX resistance of GTN patients 3.4.1. Relating to the mean (median) values of the research factor of the MTX resistant and non- resistant groups Bảng 3.5. Relating to the mean (median) values of the research factor of the MTX resistant and non- resistant groups Reseach factors Value Non- Resist MTX Resistance MTX All p Age (year) X ± SD 26,4 ± 5,0 26,2 ± 5,4 26,3 ± 5,1 0,83 Min- max 16 – 38 17 – 39 16 – 39 Latent time (month) X ± SD 2,03 ± 1,52 2,39 ± 2,52 2,13 ± 1,85 0,83 Median 2,00 1,00 2,00 0,83 Min- max 1 – 11 1 – 14 1 - 14 14 βhCG pre treatement (IU/l) X ± SD 8.690,9 ±18.648,1 19.532,9 ±37.887,9 11.667,2 ±25.773,5 0,03 Median 1.278,5 2.811,0 1.622,1 0,02 Min- max 9 - 118.534 48 - 213.180 9 - 213.180 Uterine volume (cm 3 ) X ± SD 126,03 ± 63,33 130,99 ± 64,13 127,39 ± 63,43 0,62 Median 111,55 107,40 110,48 0,67 Min- max 33,83 - 363,10 52,89 - 324,22 33,83 - 363,10 Largest tumour size (cm) X ± SD 3,3 ± 1,1 3,6 ± 1,1 3,4 ± 1,1 0,28 Min- max 1,3 – 6,9 1,8 – 5,5 1,3 - 6,9 - There was no difference in age, latent time, uterine volume, largest tumour size between resistant group and response MTX group, p> 0.05 . - Mean and median of βhCG pre treatement in resistant and response MTX group is statistically significant, p < 0.05 3.4.2. Relationship between patient's age and MTX resistance The age group <35 years old has a higher prevalence of MTX resistance than the <35 year old group, but the difference is not significant, p> 0.05. 3.4.3. Relationship between index pregnancy and MTX resistance The prevalence of MTX resistance in molar group and non-molar had no statistically significant differences, p> 0.05 3.4.4. Relationship between molar type and MTX resistance GTN patients after partial molar had higher rates of MTX resistance than those after completed molar, however the difference was not statistically significant, with p> 0.05. 3.4.5. Relationship between latent time and MTX resistance The group with latency period ≥ 7 months had higher rate of MTX resistance than the other group, but the difference was not statistically significant, p> 0.05. 3.4.6. Relationship between history of MTX treatment and MTX resistance GTN patients with a history of MTX treatment had a higher prevalence of MTX resistance, the difference was statistically significant, with p = 0.02. 3.4.7. Relationship between metastases and MTX resistance Patients with lung or vaginal metastases had a higher rate of MTT resistance than the non-metastatic group, but the difference was not significant, with p> 0.05. 15 3.4.8. Relationship between metastatic site and MTX resistance Patients with 1 to 4 metastatic nodules had a higher prevalence of MTX resistance than those without metastases, but the difference was not significant, p> 0.05. 3.4.9. Relationship between βhCG pre treatment and MTX resistance Table 3.6. Relationship between βhCG pre treatment and MTX resistance βhCG pre treatment (IU/l) MTX resistance OR (95% CI) p No Yes n % n % < 10.000 (n = 151) 116 76,8 35 23,2 2,17 (1,11 - 4,24) 0,02 ≥ 10.000 (n = 53) 32 60,4 21 39,6 The group with βhCG concentration pre treatment ≥ 10,000 IU/l had the risk of MTX resistance increased by 2.17 times compared to the group with βhCG <10,000 IU / l, p= 0.02. 3.4.10. Relationship between uterine volume and MTX resistance The group with uterine volume> 240 cm3 had higher MTX resistance rate than the other group, but the difference was not statistically significant, p> 0.05. 3.4.11. Relationship between luteal cyst and MTX resistance The group with luteal cyst had a lower prevalence of MTX resistance than the other group, but the difference was not statistically significant, p> 0.05. 3.4.12. Relationship between largest tumor size and MTX resistance Table 3.7. Relationship between largest tumor size and MTX resistance (n = 68) Largest tumor size (cm) MTX resistance OR (95% CI) p No Yes n % n % < 5 (n = 57) 42 73,7 15 26,3 4,90 (1,25 – 19,14) 0,03 ≥ 5 (n = 11) 4 36,4 7 63,6 The size of the tumor group ≥ 5 cm had the risk of MTX resistance increased 4.9 times compared to the size group <5 cm, the difference was statistically significant, p =0.03. 3.4.13. Relationship between FIGO score and MTX resistance The higher the FIGO score, the higher the MTX resistance rate. The difference between groups is statistically significant, with p = 0.02. 3.5. Evaluating Doppler ultrasound related to MTT prognosis 3.5.1. Features of Doppler ultrasound in non-resistant and MTX resistant groups 16 Table 3.8. Features of Doppler ultrasound in non-resistance and MTX resistance groups Doppler factors Value Non-resistance MTX (n = 148) MTX resistance (n = 56) p PSV (cm/s) X ± SD 60,7 ± 27,1 72,9 ± 37,0 0,03 Median 55,65 63,10 0,04 Min- max 21,4 - 214,0 23,9 - 176,0 EDV (cm/s) X ± SD 15,4 ± 19,9 25,6 ± 26,4 0,01 Median 8,25 16,20 0,02 Min- max 0 - 123,0 0 - 93,2 PI X ± SD 2,21 ± 1,13 1,93 ± 1,31 0,03 Median 2,06 1,60 0,03 Min- max 0,54 - 5,98 0,31 - 6,15 RI X ± SD 0,79 ± 0,16 0,72 ± 0,21 0,02 Median 0,83 0,76 0,02 Min- max 0,41 - 1,00 0,25 - 1,00 S/D X ± SD 6,00 ± 6,43 3,87 ± 2,50 0,03 Median 4,79 2,71 0,001 Min- max 1,70 - 59,00 1,34 - 10,93 - PSV, EDV of the MTX resistance group were higher than the non- resistance group and were statistically significant at both mean and median values, with p <0.05. - Mean value, median PI of the resistance group was lower than the non-resistant group with statistically significant, with p <0.05. 3.5.2. The ROC curve of UAPI predicts MTX resistance in GTN patients Figure 3.3. ROC curve of UAPI predict MTX resistance 0 .0 0 0 .2 5 0 .5 0 0 .7 5 1 .0 0 S e n si tiv ity 0.00 0.25 0.50 0.75 1.00 1 - Specificity Area under ROC curve = 0.5960 17 - The AUC of UAPI is 0.60. The most appropriate cut-off point is a PI value of 1.2. The sensitivity is 79.1% and the specificity is 41.1%. 3.5.3. Relationship between UAPI at threshold 1.2 and MTX resistance Table 3.9. Relationship between UAPI at threshold 1.2 and MTX resistance UAPI MTX resistance OR (95% CI) p No Yes n % n % > 1,2 (n = 148) 115 77,7 33 22,3 2,63 (1,35 – 5,11) 0,004 ≤ 1,2 (n = 56) 33 58,9 23 41,1 Patients with UAPI ≤ 1.2 had the risk of MTX resistance increased by 2.63 times compared to those with a UAPI > 1.2, with p =0.004. 3.5.4. Multivariate regression analysis of factors related to MTX resistance Table 3.10. Multivariate regression factors related to MTX resistance Yếu tố nguy cơ OR 95% CI p UAPI > 1,2 1 ≤ 1,2 5,89 1,44 – 24,13 0,014 βhCG pretreatment (IU/l) < 10.000 1 ≥ 10.000 1,50 0,45 – 4,99 0,507 Largest tumour size (cm) < 5 1 ≥ 5 4,21 0,88 – 20,04 0,071 - Through multivariate regression analysis, it shows that UAPI has significance in prognosis for treatment of MTX with p <0.05 3.6. Analysis the value of UAPI in combination with FIGO score for prognosis MTX resistance in GTN patients 3.6.1. The significance of FIGO score when combining UAPI> 1.2 Table 3.8. Significance of FIGO score when combining UAPI> 1.2 FIGO score Resistance rate to MTX FIGO alone FIGO + PI > 1,2 n Resis % n Resis % 18 ≥ 0 204 56 27,45 148 33 22,30 ≥ 1 154 50 32,47 99 27 27,27 ≥ 2 96 35 36,46 49 14 28,57 ≥ 3 53 22 41,51 22 5 22,73 ≥ 4 20 10 50,00 8 2 25,00 ≥ 5 7 3 42,86 4 0 0,00 - When using FIGO score alone, the rate of MTX resistance tends to increase with FIGO score. - In the same FIGO score group, patients with UAPI> 1.2 had a lower MTX resistance rate than those using FIGO alone. 3.6.2. The significance of FIGO score when combining UAPI ≤ 1.2 Table 3.9. Significance of FIGO score when combining UAPI ≤ 1.2 FIGO score Resistance rate to MTX FIGO alone FIGO + PI ≤ 1,2 n Resis % n Resis % ≥ 0 204 56 27,45 56 23 41,07 ≥ 1 154 50 32,47 55 23 41,82 ≥ 2 96 35 36,46 46 21 45,65 ≥ 3 53 22 41,51 31 17 54,84 ≥ 4 20 10 50,00 12 8 66,67 ≥ 5 7 3 42,86 3 3 100,00 - The MTX resistance tends to increase when using FIGO score combining UAPI ≤ 1.2. - In the same FIGO score group, patients with UAPI ≤ 1.2 had a higher MTX resistance rate than those using FIGO alone 3.6.3. FIGO score < 3 group combined threshold 1.2 UAPI (n = 151) Table 3.13. FIGO score < 3 group combined threshold 1.2 UAPI UAPI MTX resistance OR (95% CI) p No Yes n % n % > 1,2 (n = 126) 98 77,8 28 22,2 1,105 (0,40 – 3,03) 0,846 ≤ 1,2 (n =25) 19 76,0 6 24,0 19 - With FIGO score of 0 - 2, the prevalence of MTX resistance in the group with UAPI ≤ 1.2 is not much higher than that of the group having UAPI > 1.2 and the difference is not statistically significant,p> 0,05. 3.6.4. FIGO score ≥ 3 group combined threshold 1.2 UAPI (n = 53) Table 3.14. FIGO score ≥ 3 group combined threshold 1.2 UAPI UAPI MTX resistance OR (95% CI) p No Yes n % n % > 1,2 (n = 22) 17 77,3 5 22,7 4,129 (1,22 – 14,02) 0,019 ≤ 1,2 (n = 31) 14 45,2 17 54,8 - With FIGO score of 3 - 6, the prevalence of MTX resistance in the group with UAPI ≤ 1.2 is higher than that in the group with UAPI > 1.2 (54.8% compared to 22.7%) and the difference is significant, p < 0.05. CHAPTER 4: DISCUSSION 4.1. Discuss the object and research method The study subjects of 204 patients diagnosed with GTN were carefully selected according to FIGO diagnostic criteria to help reduce random errors in sample selection. 4.2. Discuss the factors related to prognosis of MTX resistance Age: results showed that there were no statistically significant differences in MTX resistance among age groups, similar to the authors Gueye, Phan Chi Thanh and Nguyen Quang Bac. Index pregnancy: our study has not found an increase of MTX resistance after abortion or stillbirth. Type of molar: Analysis of 166 GTN post molar found that the rate of MTX resistance post complet molar or partial molar was not significantly different. This result is consistent with the comments of authors such as Hemida, Gilani and Growdon. Latent time: We also found no association between latency time and MTX resistance rate. History of MTX treatment: we found that a recurrent GTN patient who had a history of MTX treatment was all resistant to the MTT chemotherapy this time and was significantly different from the group without a history of MTX treatment 20 Location and number of metastase: Our study is due to the small number of metastase and metastases in less prognostic organs (lung or vagina), so as author Roberts did not see relationship between the number of metastase and the rate of MTX resistance. The βhCG pretreatment: The higher the βhCG pretreatment, the higher the resistance to MTX chemotherapy. Our study was similar to that of domestic and foreign authors, showing that βhCG before treatment is valuable for predicting the risk of chemotherapy resistance.