Two cases are currently alive, while 54 have died. The mean survival was 4.34 months. Table 1 compares patient backgrounds in the lenvatinib, PTX, and BSC treatment groups. No significant differences were found in age, gender, weight, or TNM staging (AJCC Cancer Staging Manual 8th Edition, 2017). Additionally, no significant differences were found in the percentage of radiotherapy intervention, lung metastases, and maximum tumor size. Survival was significantly longer in the lenvatinib group, averaging 5.8 months (P = 0.004). The PTX and BSC groups survived 1.98 and 1.20 months, respectively.
Table 1
List of patient background in lenvatinib, PTX, and BSC treatment groups
|
|
Treatment
|
|
Factor
|
Group
|
BSC
|
Lenvatinib
|
PTX
|
p-value
|
N
|
8
|
36
|
12
|
|
Age
|
74.75 [62, 87]
|
72.33 [47,85]
|
72.75 [61,86]
|
0.809
|
Sex (%)
|
Female
|
5 (62.5)
|
19 (54.3)
|
7 (58.3)
|
0.904
|
|
Male
|
3 (37.5)
|
16 (45.7)
|
5 (41.7)
|
|
PS (%)
|
0
|
3 (37.5)
|
27 (75.0)
|
9 (75.0)
|
0.119
|
|
1
|
4 (50.0)
|
9 (25.0)
|
2 (16.7)
|
|
|
2
|
1 (12.5)
|
0 (0.0)
|
1 (8.3)
|
|
Stage (%)
|
ⅣB
|
1 (12.5)
|
7 (19.4)
|
1 (8.3)
|
0.634
|
|
ⅣC
|
7 (87.5)
|
29 (80.6)
|
11 (91.7)
|
|
Lung metastasis (%)
|
7 (87.5)
|
29 (80.6)
|
11 (91.7)
|
0.634
|
Radiation therapy (%)
|
0 (0.0)
|
9 (25.0)
|
1 (8.3)
|
0.155
|
Body weight (kg)
|
47.00 [40,58]
|
56.27 [41,88]
|
51.29 [37,70]
|
0.15
|
Maximum diameter(mm)
|
56.79 [37,85]
|
48.78 [24,92]
|
57.29 [30,92]
|
0.213
|
Overall survival (months)
|
1.20 [0.3,2.2]
|
5.83 [0.5,28.9]
|
1.98 [0.2,4.9]
|
0.004*
|
*P < 0.05. Continuous variables are indicated using median and range [minimum and maximum]. |
TNM staging was performed using the 8th edition of the AJCC staging system for thyroid cancer (AJCC-8). |
Table 2 shows the results of the 36 lenvatinib-treated patients with the best response. The mean starting dose and treatment duration was 20.2 mg and 4.89 months, respectively. Median PFS was 3.5 months (95% CI: 2.3–5.37). In contrast, the PTX group revealed no cases of PR, and the mean duration of treatment was 1.65 months.
Table 2
Comparison of lenvatinib and paclitaxel outcomes
Factor
|
lenvatinib
|
PTX
|
N
|
36
|
12
|
ORR
|
Partial Response (%)
|
12 (33.3)
|
0
|
|
Stable disease (%)
|
19 (52.8)
|
6 (50.0)
|
|
Progressive disease (%)
|
4 (11.1)
|
3 (25.0)
|
|
Not evaluated (%)
|
1 (2.8)
|
3 (25.0)
|
Starting dose (mg)
|
20.2 (4.7)
|
a
|
Duration of treatment (months)
|
4.89 (5.0)
|
1.65 (1.3)
|
a, 80 mg/m2 weekly |
Figure 1 shows the comparison of OS between the lenvatinib and PTX groups. The median OS was 4.77 and 2.07 months in the lenvatinib and PTX groups, respectively, indicating a significant survival benefit (P = 0.0000163).
The average lenvatinib treatment duration was 4.89 months, and Fig. 2 graphically depicts the AEs that appeared during that time. The most common AE was hypertension, which occurred in 29 (80.6%) patients, but there were no Grade 3 or higher AEs that would interfere with continued treatment. The next most common AEs were loss of appetite in 18 (50.0%), cavitation in 17 (47.2%), proteinuria and fatigue in 15 (41.7%), necrosis in 14 (38.9%), cutaneous fistula in 12 (33.3%), and tracheal fistula (including pharyngoesophageal fistula) in 9 (25.0%), and hand-foot syndrome in 8 (22.2%) patients. Necrosis was observed in 2 patients, who died of hemorrhage. Two patients had Grade 3 loss of appetite and one had a gastrointestinal hemorrhage, of whom treatment was discontinued and the patient was treated with BSC.
Surgical treatment was performed for tracheotomy upon initial presentation to avoid asphyxia or for local control in five cases in the lenvatinib group and in three cases in the other treatment groups, all of which were resectable and positive for margins. Conversion surgery was possible in two cases, and their histopathological images are shown in Fig. 3. No residual tumor was observed on gross examination in both cases; however, pathology was positive for margins. Fibrosis without necrosis was found in the tumor in case A and localized tumor necrosis was found in case B; however, both specimens showed residual viable cells. Local tumor necrosis was observed in case B; however, both specimens showed residual viable cells. Radical surgery was impossible although lenvatinib treatment reduced the tumor size.
Figure 4 shows the evolution of the maximum tumor diameter after lenvatinib treatment. The plots are shown until the final image evaluation at the end of treatment. Tumor shrinkage is observed within 1–2 months of treatment in most cases. Thereafter, treatment is maintained at the current level and discontinued at 4 months. The longest period of imaging evaluation is 9 months.
CGP results were listed in Table 3. The CGP results of 13 patients with ATC showed that 6 (46.2%) had BRAF mutation, 4 (30.8%) had RAS mutation, 7 (53.8%) had TERT mutation, 9 (69.2%) had TP53 mutation, and 1 (7.7%) had RET fusion, TMB high, PTEN and FGF mutation, respectively. RET fusion, TMB high, PTEN, and FGF mutations were detected in one case (7.7%), respectively. Among the 10 patients with papillary thyroid cancer (PTC), 10 (100.0%) had BRAF mutation, 7 (70.0%) had TERT mutation, and 1 (70.0%) had high TMB. Among the four patients with follicular thyroid cancer (FTC), one had a RAS mutation, but none had a TP53 mutation. Therefore, ATC cases 1–7 are of PTC origin, while cases 8–12 are of FTC origin; cases 1–3 are consistent with the coexistence of PTC in the surgical specimens, and case 7 is consistent with PTC as the tissue before the anaplastic transformation.
Table 3A. Results of genetic analysis at anaplastic thyroid cancer
Table 3B. Genetic analysis results of patients with differentiated thyroid cancer who have completed lenvatinib treatment