4. Discussion
Chemotherapy for lung cancer has made remarkable progress, and its drug selection has been subdivided by genetic analysis. Although RECIST is an international, highly reproducible method used to assess the response to chemotherapy and/or radiation therapy for neoplasms [
6], the RECIST criteria have limitations, the RECIST criteria have limitations, especially in assessing whether newer cancer therapies stabilize the disease [
7]. RECIST has the following problems: (1) Only maximum diameters of tumors are used for analysis: (2) There is no metabolic or functional information; (3) It does not include the volume of the neoplasm: (4) CT values and cavitary lesions are not included.: (5) Partial solid nodules are evaluated with lung window.
Oxnard et al. analyzed the intraobserver variability of maximum diameter measurements in 30 non-small cell lung cancer patients who underwent repeat CT imaging within 15 minutes and reported that 6% of lesions were erroneously determined to be PD, especially for lesions less than 3 cm in diameter, suggesting caution in the measurement of small lesions [
8]. Zhao et al. also reported that inter- and intra-observer variability in size measurement by radiologists was greater than that by computerized measurement, and that they overestimated PD, especially when the size was small [
9]. And it has been pointed out that RECIST has the disadvantage that it cannot accurately evaluate the therapeutic effect of molecularly targeted drugs and immune checkpoint inhibitors in patients with lung cancer.
On the other hand, recent advances in 3D workstations have made accurate and automated volumetry possible, and there have been reports of less inter- and intra-observer variability [
10,
11,
12] and better prognostic correlation with volumetric change rates than with RECIST [
13].
In 2004, Takahara et al. introduced diffusion-weighted whole-body imaging with background suppression (DWIBS) technology [
14]. The MR signal measured by DWI is attenuated by the random molecular motion (Brownian motion) of water molecules. Tumors with high cell density have a narrower extracellular space, restricting the movement of protons within the tumor, resulting in a lower apparent diffusion coefficient (ADC) compared to normal tissue. Diffusion-weighted imaging (DWI) can be used to differentiate benign from malignant lesions in the lung, thorax, prostate, breast, and liver. Whole-body DWI (DWIBS) is also currently used for cancer diagnosis and staging, and for determining the effectiveness of treatment for prostate and breast cancer. [
14] [
15].
FDG-PET-CT is often used to determine the efficacy of treatment for lung cancer, but has drawbacks compared to DWIBS, such as radiation exposure and cost effectiveness. Although DWI was reported to be superior to CT in determining the efficacy of chemotherapy and radiation therapy for recurrent lung cancer tumors, the utility of DWIBS is not yet widely known to respiratory physicians and thoracic surgeons. [
16]. On the other hand, several studies have reported that the diagnostic efficacy of DWIBS is comparable to that of FDG-PET/CT [
17,
18,
19]. DWI has been reported to be more accurate than FDG-PET/CT in detecting primary tumors and evaluating lymph nodes in non-small cell lung cancer (NSCLC) [
20]. DWI has better diagnostic capabilities than PET in the evaluation of pulmonary nodules and masses [
21]. DWI using ADC values and signal intensity can be useful for differential diagnosis of mediastinal lymph nodes [
22]. The superiority of DWI can be explained by the fact that DWI not only has fewer false positives [
23], but also fewer false negatives in the N stage of NSCLC compared to FDG-PET/CT [
20]. DWI is useful in differentiating pleural effusion from pleural effusion with pleural dissemination or malignant pleural mesothelioma [
24].
We compared the ability of FDG-PET to detect primary tumor, regional lymph node metastasis, and distant metastasis in 40 patients who had undergone DWIBS between January 2014 and September 2020 at our hospital for close examination of lung cancer and FDG-PET/CT within 3 months before and after the FDG-PET/CT. Of all lesions with FDG-PET accumulation, 97.5% (39/40) were also depicted by DWIBS for the primary tumor, 81.4% (35/43) for regional lymph node metastases, and 87.5% (35/40) for distant metastases.
DWIBS had almost the same imaging performance as FDG-PET for the primary tumor. As for the regional lymph node metastasis, in the cases that showed accumulation on FDG-PET and tested negative on DWI, the nodal signal was lower on DWI, which seemed to pick up the lesion more difficult than on FDG-PET. The signal on DWI may have been reduced by irregular breathing and heartbeat. On the other hand, it is possible that the FDG-PET results were false positive. As for distant metastasis, FDG-PET did not show accumulation in the area indicated as bone metastasis by DWIBS in 5 cases, and DWIBS failed to delineate a microscopic nodule that appeared to be pleural dissemination in 1 case. In one of these cases, DWIBS imaging one month after chemotherapy was initiated showed decreased signal of bone lesions and the appearance of bone sclerosis on subsequent CT, suggesting a high likelihood of bone metastases. The remaining cases were probably false positives because there were no changes after treatment. Thus, DWIBS can be used as an alternative to FDG-PET because there is no extreme difference in lesion delineation compared to FDG-PET.
ADC in DWI may be useful in monitoring early response to chemotherapy and predicting prognosis in non-small cell lung cancer (NSCLC) [
25] [
26]. The higher mean pretreatment ADC of liver metastatic lesions of colorectal cancer predicted a poorer response to chemotherapy [
27]. In most of the published studies, there was a significant correlation between change in ADC values and treatment response, with ADC increasing in most cases of response [
28]. This study also showed a tendency for the rate of change in tDV to decrease as the rate of change in ADC increased in the positive direction.
We examined the change of total diffusion volume (tDV) and compared the treatment effect with RECIST and with DWIBS using BD score. A tDV change of about 40% seemed to be a good criterion to judge PR and PD. Compared RECIST on CT, DWIBS using BD score would be more accurate for response evaluation of treatment. The patients downgraded using DWI included those with small hilar lung cancer or hilar lymph node metastases. The DWI underestimation may have been due to poor visualization caused by air or heartbeat.
The flare phenomenon or pseudoprogression seen after radiation therapy or chemotherapy has been known for some time and is particularly problematic in differentiating it from true tumor progression, which is often difficult to make a definitive diagnosis with a single imaging study. The mechanism is that immunotherapy activates T lymphocytes, which accumulate in the tumor and seemingly increase the size of the tumor, according to pathological reports [
29]. Misjudging a true tumor exacerbation may shorten the prognosis by changing an effective therapeutic agent to another agent.
Tazdait et al. reported that among 160 non-small cell lung cancer patients treated with immune checkpoint inhibitors, 20 (13%) had atypical responses, including pseudoprogressions in 8 (5%) and dissociated responses in 12 (8%) patients [
30]. Dissociated responses is one of the atypical immune-related response patterns and refers to cases of opposite responses that differ between organs. In the present study, there was a tendency for tDV to decrease with increasing ADC values, but some of the cases with increasing ADC values and increasing tDV appeared to have dissociated responses. Therefore, when ADC values increase and tDV increases, it is necessary to evaluate with the possibility of dissociated response or pseudoprogression in mind. DWIBS can detect multiple metastatic lesions throughout the entire body and differentiate malignancy from benignity in only one examination. DWIBS has benefits of diagnostic accuracy and is less expensive in medical costs for the detection of a relapse. DWIBS could potentially replace FDG-PET/CT after lung cancer resection [
31].
For usefulness of BD score, Usuda et al. reported that ADC histograms using BD score could discriminate pulmonary abscesses and mycobacterial infections (PAMIs) from lung cancers, which discrimination was usually difficult in CT, PET-CT or MRI. ADC histogram has the potential to be a valuable tool to differentiate benignity from malignancy [
32].
This study has several limitations. This study was performed at a single institution and dealt with a small number of patients. We were unable to locate many cases that had undergone 2 examinations of CT and MRI in the period from August 2014 to August 2020. So our sample size is limited.