Table 3.
Clinical application of MRI in cancer-targeted therapy
| Technique | Drug | Target | Tumor | N | Results | References |
|---|---|---|---|---|---|---|
| DCE-MRI | Bevacizumab | VEGF | Breast cancer | 70 | Significant decreases in Ktrans, kep, Ve, and AUC60 after cycle 5 of treatment. | 291PMID: 34298725 |
| DCE-MRI | Bevacizumab (+chemotherapy) | VEGF | Breast cancer | 19 | Significant decreases in Ktrans, kep, and IAUGC at 180 s after cycle 1 of treatment. The median relative change in the slope of the wash-in curve from baseline to cycle 4 was significantly different between responders and nonresponders. | 364PMID: 17709827 |
| DCE-MRI | Bevacizumab (+chemotherapy) | VEGF | Breast cancer | 21 | Decreases in Ktrans, kep, and Ve after cycle 1 of treatment. No correlation with treatment response. | 290PMID: 16391297 |
| DSC-MRI | Bevacizumab (+radiotherapy) | VEGF | Glioblastoma | 67 | OS benefit from bevacizumab plus radiotherapy compared with radiotherapy alone was observed for larger baseline MRI contrast-enhancing tumors and for higher ADC. | 300PMID: 32967939 |
| DSC-MRI | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 123 | Quantitative DT1 showed a significant difference in OS at week 8 between responders and nonresponders/nonprogressors. | 365PMID: 31248863 |
| DSC-MRI | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 254 | Decreases in nrCBV, nrCBF, and nTMRO2 values after bevacizumab treatment. None of these parameters was predictive of OS. | 302PMID: 32720870 |
| DSC-MRI | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 21 | Early decreases in rCBV were predictive of improved survival. | 303PMID: 25646027 |
| DSC-MRI and DCE-MRI | Bevacizumab (+chemoradiation therapy) | VEGF | Glioblastoma | 42 | High pretreatment rCBV was predictive of improved OS. | 305PMID: 32678438 |
| DSC-MRI and DCE-MRI | Bevacizumab (+chemotherapy) | VEGF | Breast cancer | 22 | A lower ΔKtrans or ΔADC reduction in 21 days after treatment predicted shorter CNS-specific PFS. A lower ΔPeak or ΔIAUC60 reduction predicted shorter OS. | 295PMID: 29770848 |
| DSC-MRI and DCE-MRI | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 33 | PFS increased significantly with time to the maximum value of the residue (Tmax). OS decreased significantly with srCBV and increased significantly with Tmax. | 366PMID: 33828310 |
| MRS | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 13 | Increased NAA/Cho at 8 weeks and decreased Cho/Cr and increased NAA/Cr and NAA/Cho at 16 weeks post-treatment was associated with both 6-month progression-free survival and 1-year survival. | 317PMID: 23645534 |
| MRS | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 21 | A lower mI/c-Cr in the intratumoral and peritumoral volume before and during treatment was predictive of poor survival. | 367PMID: 34751617 |
| VHL and VAM | Bevacizumab | VEGF | Glioblastoma | 13 | Early response to bevacizumab was dominated by the reduction of smaller microvasculature. | 326PMID: 28819189 |
| DWI | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 123 | High pretreatment contrast-enhancing tumor volume was associated with shorter PFS and OS. A high volume fraction of increasing ADC after therapy was associated with shorter PFS, while a high volume fraction of decreasing ADC was associated with shorter OS. | 301PMID: 25672376 |
| DWI | Bevacizumab (+chemotherapy) | VEGF | Colorectal liver metastasis | 74 | Post-treatment ADCmean was significantly associated with OS and PFS. | 312PMID: 35013857 |
| DWI | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 32 | Pretreatment tumor volume was correlated with OS. Patients with high ADCL had favorable survival when treated with bevacizumab. | 313PMID: 32365185 |
| DWI | Bevacizumab (+chemotherapy) | VEGF | Glioblastoma | 242 | ADClow was an independent prognostic parameter for OS and PFS. Patients with ADClow ≥ 1241 × 10−6 mm²/s had prolonged OS compared with those with ADClow < 1241 × 10−6 mm²/s. | 314PMID: 32393964 |
| DCE-MRI and DWI | Bevacizumab (+chemotherapy) | VEGF | Colorectal liver metastasis | 126 | D-RECIST- but not RECIST-defined responders had significantly longer median DFS than nonresponders. D-RECIST- but not RECIST-defined responses independently predicted DFS. | 310PMID: 33449175 |
| APT MRI and DWI | Bevacizumab | VEGF | Glioblastoma | 54 | Mean APT signal intensity change after bevacizumab treatment indicated a low 12-month progression rate and longer PFS. High mean normalized CBV at follow-up was associated with a high 12-month progression rate and shorter PFS. Mean APT signal intensity change was a significant predictor of diffuse non-enhancing progression, whereas follow-up 95th percentile of the normalized CBV was a predictor of local enhancing progression. | 323PMID: 32154775 |
| CEST- EPI | Bevacizumab (with or without adjuvant chemotherapy or immunotherapy) | VEGF | Glioblastoma | 11 | The reduction in tumor acidity was linearly correlated with PFS, being a significant predictor of PFS. | 322PMID: 30806888 |
| DCE-MRI and FLAIR | Bevacizumab (+chemoradiation therapy) | VEGF | Glioblastoma | 159 | Increasing 2D-T1 and FLAIR post-treatment significantly predicted worse OS. Adjusting for 2D-T1 and treatment, increasing FLAIR represented a significantly higher risk for death. | 368PMID: 29590461 |
| FLAIR | Bevacizumab (with or without chemotherapy) | VEGF | Gliomas | 33 | Lower edge contrast of the FLAIR hyperintense region was associated with poorer PFS and OS. | 369PMID: 29622553 |
| DCE-MRI and FLAIR | Bevacizumab (with or without chemotherapy) | VEGF | Glioblastoma | 119 | Early MRI response could predict PFS and OS. Early MRI progression was a strong independent predictor of mortality. | 370PMID: 28678383 |
| T2WI and DCE-MRI | Bevacizumab (+chemoradiation therapy) | VEGF | Glioblastoma | 232 | At weeks 6 and 12 of treatment, increases in baseline necrosis and de novo necrosis were strongly associated with worse OS and PFS. | 371PMID: 31076534 |
| TME mapping | Bevacizumab | VEGF | Glioblastoma | 18 | Higher percentage of neovascularization and active tumor in baseline indicated poor or no treatment response. | 329PMID: 30361791 |
| PWI | Angiocept, bevacizumab, cilengitide, enzastaurin, sorafenib, thalidomide and vandetani | VEGF | Glioblastoma | 117 | Patients with an angiogenic subtype of glioblastoma benefited from antiangiogenic therapy with improved OS. | 372PMID: 28007759 |
| DCE-MRI and DWI | Bevacizumab or aflibercept or cediranib or cabozantinib | VEGF | Glioblastoma | 258 | Baseline ADCL was an independent predictive biomarker for OS in anti-VEGF therapies. An ADCL threshold of 1.24 μm2/ms produced the largest OS differences between patients. | 309PMID: 28655794 |
| DWI-MRI | Lenvatinib and toripalimab | VEGF | Intrahepatic cholangiocarcinomas | 43 | ADC was an independent variable associated with early progression. Patients with low ADC values showed shorter PFS. | 307PMID: 35488518 |
| Gd-EOB-DTPA-enhanced MRI | Lenvatinib or atezolizumab and bevacizumab | VEGF | Hepatocellular carcinoma | 68 | No predictive association between PFS and EOB-MRI in the lenvatinib group. In the atezolizumab plus bevacizumab group, the heterogeneous type and hyperintensity type had significantly shorter PFS than the homogeneous type and the hypointensity type, respectively. | 332PMID: 35159095 |
| DSC-MRI and DCE-MRI | Cabozantinib | VEGF | Glioblastoma | 108 | A log-linear association between baseline tumor volume and OS and a linear correlation between initial change in tumor volume and OS were observed. Continuous measures of baseline tumor volume and volumetric response were independent predictors of OS. | 299PMID: 29660005 |
| DSC-MRI and DCE-MRI | Lenalidomide or axitinib | VEGF | Hepatocellular carcinoma | 74 | Greater reductions in ΔPeak or ΔAUC on days 3 and 14, and ΔKtrans on day 14 were associated with better PFS. Greater reductions in ΔAUC or ΔKtrans on day 14 were associated with better OS. ΔKtrans on day 14 was an independent predictor of PFS after controlling for ORR and DCR. | 296PMID: 34638446 |
| ASL MRI | Sunitinib or pazopanib | VEGF | Renal cell carcinoma | 28 | Responders had higher baseline tumor perfusion than nonresponders. Interval reductions in perfusion at week 2, cycle 2, and cycle 4 were not associated with ORR or PFS. | 327PMID: 33258745 |
| DWI | Sunitinib, pazopanib or axitinib | VEGF | Renal cell carcinoma | 92 | Patients with >5 bone metastases (BM) on WB-DWI/MRI had a lower response rate, and more frequently suffered early progressive disease, shorter PFS, and shorter OS than patients with ≤5 BM. | 373PMID: 32297532 |
| DWI | Sunitinib | VEGF | Gastrointestinal stromal tumor | 15 | Pretreatment β and ΔD differed between good- and poor-responding lesions. Combining ΔD with pretreatment β obtained an improved AUC (0.843) with a predictive accuracy of 75.7%. | 308PMID: 28643387 |
| DCE-MRI and FLAIR | Sunitinib | VEGF | Renal cell carcinoma | 34 | Higher baseline and day 14 values for Ktrans were significantly associated with longer PFS. | 374PMID: 29383520 |
| DWI | Imatinib or sunitinib | VEGF | Gastrointestinal stromal tumor | 62 | The percentage change of ADC and longest diameter after 2 weeks of therapy were significantly different between responders and nonresponders. | 306PMID: 30103713 |
| DCE-MRI | Regorafenib | VEGF | Colorectal cancer | 27 | >70% drop in KEF (6/23) was associated with a higher disease control rate at 2 months and improved PFS and OS. | 297PMID: 28790159 |
| DCE-MRI | Sorafenib | VEGF | Hepatocellular carcinoma | 29 | Stratification according to mRECIST and vqEASL successfully captured response and stratified OS, while stratification according to RECIST and %qEASL did not correlate with OS. | 298PMID: 33123796 |
| Gd-EOB-DTPA-enhanced MRI | Sorafenib (with or without selective internal radiation therapy) | VEGF | Hepatocellular carcinoma | 312 | High gadoxetic acid uptake on pretreatment MRI was significantly associated with shorter OS. | 375PMID: 34541612 |
| Gd-EOB-DTPA-enhanced MRI | Sorafenib (with or without selective internal radiation therapy) | VEGF | Hepatocellular carcinoma | 376 | Peritumoral arterial enhancement and peritumoral hypointensity in hepatobiliary phase were predictors of worse OS. Peritumoral hypointensity in hepatobiliary phase was a predictor of liver decompensation. | 376PMID: 34686780 |
| Gd-EOB-DTPA-enhanced MRI | Sorafenib | VEGF | Hepatocellular carcinoma | 65 | Regular tumor margin and the presence of tumor thrombus were indicators of high RAF1 expression. | 377PMID: 34738148 |
| Gd-EOB-DTPA-enhanced MRI | Sorafenib | VEGF | Hepatocellular carcinoma | 91 | The presence of incomplete capsule or intratumoral vessels and the absence of capsule were potential indicators of high BRAF and RAF1 expression. | 378PMID: 30547202 |
| MRE | Sorafenib | VEGF | Hepatocellular carcinoma | 50 | Higher MRE-assessed liver stiffness was significantly associated with poor OS and significant liver injury after sorafenib therapy. | 328PMID: 33033862 |
| MRS | Cediranib (+chemoradiation therapy) | VEFG | Glioblastoma | 40 | Total Cho/healthy Cr after 1 month of treatment was significantly associated with OS. | 379PMID: 29202103 |
| IVIM-MRI and DCE-MRI | Lenalidomide | VEGF | Hepatocellular carcinoma | 44 | Participants with a higher slope, Kep and ADC values had longer PFS. Participants with small tumor size, higher slope, ADC and f values had longer OS. Kep and ADC were independent predictors of PFS. Slope and ADC were independent predictors of OS. | 325PMID: 34441274 |
| DCE-MRI | Bevacizumab and erlotinib | VEGF and EGFR | NSCLC | 44 | Whole-tumor Ktrans was not associated with PFS, but patients with an increase of more than 15% in the SD of tumor Ktrans values after 3 weeks had shorter PFS. | 200PMID: 21149474 |
| DWI | Bevacizumab and erlotinib (+chemoradiation therapy) | VEGF and EGFR | Glioblastoma | 36 | A lower ADC percentile value within the T2-hyperintense lesion (T2L) at early follow-up timepoints was associated with worse outcomes. The ADC10% within the T2L at 2 months was strongly associated with OS and PFS. | 315PMID: 25351579 |
| DCE-MRI and DWI | Gefitinib (+radiotherapy) | EGFR | Nonsmall-cell lung cancer | 253 | Tumor regression rate, ADCpost, ΔADCpost, and ΔADCpost (%) were key imaging indicators for predicting the outcome. | 311PMID: 34514171 |
| DCE-MRI | Trastuzumab or T-DM1 ( + chemotherapy) | HER2 | Breast cancer | 46 | Interim changes in ETV value were highly correlated with residual cancer burden. | 380PMID: 29641224 |
| DCE-MRI | Trastuzumab or/and pertuzumab | HER2 | Breast cancer | 21 | Concentric tumor shrinkage pattern after targeted therapy was associated with pCR. No association between the initial enhancement ratio and pCR. | 294PMID: 31444111 |
| DSC-MRI | Trastuzumab (+chemotherapy) | HER2 | Breast cancer | 296 | Patients with early rCR on MRI achieved pCR in 73% of HER2-positive breast cancer cases and 88% in the HR-negative subgroup. Achieving rCR was associated with a rate of the 5-year recurrence-free interval of 88%, compared with 68% without rCR. | 381PMID: 28432515 |
| IVIM-MRI | Nivolumab or pembrolizumab | PD-1 | Non-small cell lung cancer (NSCLC) | 20 | An increased ADC at 8 weeks and decreased ADCkurt and ΔADCkurt 4 weeks after treatment were associated with objective responses and longer PFS. A decreased ΔADCskew at 4 weeks was associated with objective responses, disease control, and longer PFS and OS. | 324PMID: 32203770 |
| Gd-EOB-DTPA-enhanced MRI | Anti-PD-1/PD-L1 monotherapy | PD-1/PD-L1 | Hepatocellular carcinoma (HCC) | 18 | The TTnP and median PFS in HCC patients with hyperintense nodules were significantly shorter than in those with hypointense HCC nodules after treatment. | 331PMID: 34950184 |