1000/1000
Hot
Most Recent
| Version | Summary | Created by | Modification | Content Size | Created at | Operation |
|---|---|---|---|---|---|---|
| 1 | Toshimitsu Yamaoka | + 2284 word(s) | 2284 | 2021-02-24 05:15:59 | | | |
| 2 | Nicole Yin | Meta information modification | 2284 | 2021-03-09 06:58:58 | | |
Non-small-cell lung cancer patients who responded to EGFR-tyrosine kinase inhibitors (EGFR-TKIs) and obtained survival benefits had somatic EGFR mutations. EGFR-TKI-related adverse events (AEs) are usually tolerable and manageable, although serious AEs, such as interstitial lung disease (ILD), occur infrequently. The etiopathogenesis of EGFR-TKI-induced ILD remains unknown. Herein, we evaluated the relationship between EGFR-TKIs and ILD. We discussed the relevance of cytotoxic agents or immunotherapeutic agents in combination with EGFR-TKIs as a potential mechanism of EGFR-TKI-related lung injury and reviewed recent developments in diagnostics and therapeutics that facilitate recovery from ILD.
With EGFR-TKI treatment, NSCLC patients who have EGFR-activating mutations can achieve longer progression-free survival (PFS) and a higher quality of life[1][2][3]. The EGFR-TKIs are generally well tolerated because of a favorable toxicity profile compared to chemotherapeutic agents. EGFR plays an essential role in epithelial maintenance and, therefore, EGFR-TKIs might impair epithelial cell growth and migration and alter cytokine expression, leading to the recruitment of inflammatory cells and consequent tissue injury. For example, diarrhea, acneiform skin rash, and paronychia are the commonest side effects of EGFR-TKIs. Mucositis, stomatitis, corneal erosion, and epistaxis are less common, but are clinically important. These side effects are associated with the effect of wild-type-EGFR inhibition. Hepatic and pulmonary toxicity has been reported to be an EGFR-TKI-associated fatal event[4][5]. Severe hepatic dysfunction can be managed by switching to another EGFR-TKI, whereas lung injury (also known as interstitial lung disease (ILD)) often leads to the discontinuation of EGFR-TKI treatment in the affected patients[6]. ILD is a rare but serious complication of EGFR-TKI because one-third of the patients with EGFR-TKI-associated lung injury die, even after receiving intensive supportive care, including supplemental oxygen, empirical antibiotics, systemic corticosteroids, and mechanical ventilation. Therefore, lung injury is the most life-threatening adverse event (AE) related to EGFR-TKI treatment. However, the mechanisms mediating the association between EGFR-TKI treatment and lung injury is not well known. The NSCLC patients with EGFR-TKI-induced lung injury had characteristic clinical profiles of risk factors, including smoking history, male sex, and pre-existing lung fibrosis. On the basis of these patient profiles, it is possible to speculate that lung injury would be worsened by EGFR-TKI treatment in patients with chronic inflammation.
Here, we present the results of meta-analysis and clinical trials reported so far with regard to the incidence of ILD when there is: (1) EGFR-TKI monotherapy, (2) a combination of EGFR-TKIs and other chemotherapeutic agents, and (3) a combination of EGFR-TKIs and immune checkpoint inhibitors (ICIs).
Until now, various EGFR-TKI clinical trials have been conducted, and AEs have been investigated in detail. Meta-analyses that included many clinical studies have reported the occurrence of ILD[7][8][9]. The databases of these reports are partially duplicated, as shown in Table 1; however, when racial differences are not taken into account, the incidence of all grades of ILD with EGFR-TKI monotherapy is generally consistent at 1.1%–2.2%. The incidence of grade 3 or higher ILD (severe symptoms or oxygen supplemental cases) is 0.6%–1.0% and of grade 5 (mortality) AE is 0.2%–0.5%. ILD is the most frequent cause of EGFR-TKI-treatment-related death, accounting for 58% of cases[7]. There are no reports of a significant difference in the incidence of ILD between EGFR-TKIs (gefitinib 1.3%–2.2%, erlotinib 0.6%–1.5%, afatinib 0.2%–0.6%, and osimertinib 3.0%)[7][8][9].
With regard to racial differences, the incidence of ILD is especially high in the data from Japan. Suh et al. compared research data from Japan and other countries and found that the ILD incidence in Japanese patients was significantly higher, 4.8% vs. 0.6% (p < 0.001) for all grades, 2.5% vs. 0.4% (p < 0.001) for grade ≥3, and 1.0% vs. 0.2% for grade 5 AEs (p < 0.001)[10]. In contrast, when comparing Asian countries, excluding Japanese participants with non-Asians, no significant difference was observed. The reason for the high incidence of ILD in the Japanese population is being analyzed with regard to proteomic[11] and genetic polymorphisms (NEJ022A; UMIN ID: UMIN000015612), although an underlying cause has not been clarified until now. As mentioned earlier, after the occurrence of gefitinib-induced ILD emerged as a social problem in Japan and was evaluated in many clinical studies in Japan as compared to that in other countries, Japanese oncologists have specifically focused on the AE of ILD, and the high incidence may be partly attributable to the high frequency of CT examinations (the number of Japanese CTs per million people is the highest worldwide[12]. Moreover, the lack of quantitative and qualitative international criteria for IP may be a factor, although the diagnostic guideline for idiopathic pulmonary fibrosis (IPF), which represents a collaborative effort between American Thoracic Society (ATS)/European Respiratory Society (ERS)/ Japanese Respiratory Society (JRS)/Latin America Thoracic Association (ALAT), has been issued in 2018[13].
Table 1. Incidence of interstitial lung disease (ILD) on EGFR-TKI treatment in meta-analysis.
|
Author |
Patients |
Treatments |
ILD Incidence (%) |
Japanese ILD Incidence (%) |
Reference |
|
Suh et al. |
15,713 |
Gef. 1, Erlo. 2, Afa. 3, Osim. 4 |
1.1 |
4.8 |
[10] |
|
Shi et al. |
8609 |
Gef. 1, Erlo. 2 |
1.2 |
3.3 |
[14] |
|
Takeda et al. |
1468 |
Gef. 1, Erlo. 2, Afa. 3 |
0.6–2.2 |
3.8 |
[15] |
1 Gef.: gefitinib, 2 Erlo.: erlotinib, 3 Afa.: afatinib, 4 Osim.: osimertinib.
Clinical trial results have been reported on the incidence of ILD when EGFR-TKIs are used in combination with other chemotherapeutic drugs. Hosomi et al. reported that the incidence of ILD in patients receiving carboplatin (CBDCA) + pemetrexed (PEM) + gefitinib was 6% for all grades and 2.4% for grade ≥3 AEs, and the incidence of AEs with gefitinib monotherapy was 3.5% for all grades and 0.6% for grade ≥3 AEs in Japan (NEJ009)[16]. With the determination of the incidence of IP by PEM monotherapy in NSCLC patients at 2.6% in Japan, the combined use of EGFR-TKI and CBDCA + PEM may increase the incidence of ILD compared to EGFR-TKI monotherapy. In contrast, Noronha et al. similarly compared the AE incidence in CBDCA + PEM + gefitinib-administered patients with those of the gefitinib-alone group and found no significant difference in ILD incidence[17].
The incidence of ILD in erlotinib + bevacizumab[18][19][20] and erlotinib + ramucirumab[21] was investigated in patients receiving combination therapy with angiogenesis inhibitors and EGFR-TKIs. None of the patients showed a significant increase in ILD incidence compared to that with erlotinib monotherapy.
The combination of ICIs with chemotherapeutic drugs has been shown to be effective. In this context, a phase I clinical trial of the anti-programmed death-ligand-1 antibody, durvalumab, and osimertinib combination was conducted[6][22]. When osimertinib was combined with durvalumab 3 or 10 mg/kg, the incidence of ILD was as high as 22% for all grades and 8.7% for grade ≥3 AEs; however, due to the increasing reports of ILD, the clinical trial (TATTON) was terminated. The trial enrolled a relatively small number of patients; however, we infer that combination therapy with EGFR-TKIs and ICIs may increase the incidence of ILD, and development of the combination therapy regimens should be carefully evaluated (Table 2).
Table 2. Incidence of ILD in the combined therapy with EGFR-TKIs.
|
|
|
ILD |
|
|
|
Trial |
Drugs |
ALL (%) |
≥G3 (%) |
Reference |
|
NEJ009 |
CBDCA 1 + PEM 2 + Gefitinib |
11/170 (6.5%) |
4/170 (2.4%) |
[16] |
|
|
Gefitinib |
6/171 (3.5%) |
1/171 (0.6%) |
|
|
NEJ026 |
Erlotinib + Bevacizumab |
0/112 (0%) |
0/121 (0%) |
[18] |
|
|
Erlotinib |
5/114 (4.4%) |
0/114 (0%) |
|
|
RELAY |
Erlotinib + Ramucirumab |
3/221 (1.4%) |
1/221 (0.5%) |
[21] |
|
|
Erlotinib |
4/225 (1.8%) |
2/225 (0.9%) |
|
|
TATTON |
Osimertinib + Durvalumab |
5/23 (22%) |
2/23 (8.7%) |
[22] |
1 CBDCA: carboplatin, 2 PEM: pemetrexed.
EGFR-TKI-related lung injury, otherwise known as ILD, induced by EGFR-TKI, is a rare but fatal AE. The diagnosis is based on clinical suspicion during EGFR-TKI treatment, detection of lung parenchymal infiltration on radiological assessment, and diagnostic exclusion of tumor progression, cardiac diseases, or other pulmonary complications, such as infectious pneumonitis[23][24].
The clinical symptoms of EGFR-TKI-induced ILD include cough, fever, exertional dyspnea, and other respiratory symptoms. As these are non-specific symptoms, and the symptom onset occurs in a wide range of conditions, it is difficult to distinguish ILD from other respiratory diseases. Furthermore, the findings on physical examination are non-specific, although chest auscultation and consideration of other findings, such as peripheral edema, could facilitate the exclusion of a diagnosis of cardiac pulmonary edema[23][24].
In general, chest roentgenography is the first step in the radiological assessment of pulmonary diseases, including ILD. However, the findings are nonspecific and sometimes almost normal in early-stage disease. High-resolution CT (HRCT) of the chest is used in the second step of a diagnostic workup of ILD[23][24], and the findings provide information about not only the distribution pattern but also the abnormalities of the parenchymal pulmonary pattern, which is possibly correlated with the histologic pattern of ILD[25]. Endo et al. analyzed imaging data and clinical manifestations of cases of acute ILD caused by gefitinib, and proposed four radiological patterns based on chest roentgenogram and HRCT (Table 3)[26]. Furthermore, this report showed the incidence of these four patterns that correspond to ILD patterns that reflect a higher mortality risk[26]. An HRCT of the chest should be obtained before the initiation of EGFR-TKI therapy because pre-existing pulmonary abnormalities, such as lung fibrosis, are associated with an increased risk of ILD[27]. Serial assessment of HRCT during EGFR-TKI treatment is helpful for the early detection of ILD-related changes. However, HRCT assessment is not comprehensive because the findings on HRCT of EGFR-TKI-induced ILD are similar to those of other pulmonary complications, such as Pneumocystis jirovecii pneumonitis (PCP), cytomegalovirus, and coronavirus disease 2019 (COVID-19) infection[28]. Additionally, patients receiving EGFR-TKIs could be immunocompromised and have a higher risk of infection because of advanced cancer, which can lead to a wrong diagnosis.
Laboratory tests, including complete blood cell counts and biochemical analysis, have been routinely conducted in patients suspected to be at risk for developing ILD. Krebs von den Lungen-6 (KL-6) is a high-molecular weight glycoprotein that is classified as human MUC1 mucin[29]. Elevation of serum KL-6 levels is significantly correlated with various lung diseases such as lung fibrosis and radiation pneumonitis[29][30]. However, evaluating the absolute level of KL-6 in serum is insufficient to diagnose drug-induced ILD[31]. Kawase et al. demonstrated that the degree of elevation of KL-6 from baseline could predict the disease severity as well as the clinical prognosis[32]. Therefore, it is essential to evaluate serum KL-6 levels before the initiation of EGFR-TKI treatment, and close monitoring is necessary during the therapy.
Bronchoscopy is useful to obtain direct information about lung parenchymal abnormalities. However, the main aim of the examination is to exclude infectious pneumonitis and disease progression, and not to confirm the pattern of ILD[33].
Table 3. Four proposed radiological patterns based on chest roentgenogram and high-resolution CT (HRCT) on acute ILD caused by gefitinib.
|
Pattern |
Findings on Roentgenography |
Manifestations on CT Scan |
Corresponding |
Incidence (%) |
Mortality (%) |
|
A |
Diffuse and faint opacity without volume loss |
Non-specific area: Ground-glass opacity |
NSIP 1 |
47.1 |
31.0 |
|
B |
Peripheral consolidation |
Multifocal area: Airspace consolidation |
OP/BOOP 2 |
13.7 |
28.6 |
|
C |
Patchy or diffuse faint, liner opacities |
Patchy distribution area: Ground-glass opacity interlobular septal thickening |
AEP 3 |
2.0 |
0.0 |
|
D |
Diffuse faint opacity or consolidation with volume loss |
Extensive bilateral area: Ground-glass opacities; air space consolidation with traction bronchiectasis |
AIP 4 |
23.5 |
75.0 |
|
Others |
Non-specific |
Non-specific |
N/A |
13.7 |
45.5 |
1 NSIP: nonspecific interstitial pneumonia, 2 OP/BOOP: organizing pneumonia/bronchiolitis obliterans organizing pneumonia, 3 AEP: acute eosinophilic pneumonia, 4 AIP: acute interstitial pneumonia.
These are no standard guidelines, and no specific therapies have been established for the management of EGFR-TKI-related ILD. Withdrawal of EGFR-TKI treatment is essential in all cases, and supportive treatments with oxygen supplementation are required in patients with respiratory failure. Corticosteroids have been widely used in patients with ILD to control excessive pulmonary inflammation[23][24]. Treatment combining EGFR-TKIs with corticosteroids was successful among patients with NSCLC who experienced ILD induced by EGFR-TKIs[34]. However, such cases were restricted in mild ILD, had non-diffuse alveolar damage (DAD) patterns, and responded well to corticosteroid therapy. On the other hand, it was reported that cases with EGFR-TKI-induced ILD that did not respond to a moderate dose of corticosteroid could have improved with high-dose corticosteroid therapy[35]. High-dose corticosteroid therapy consisted of 500 mg/day, up to 1 g/day, of intravenous methylprednisolone for 3 days. Thereafter, corticosteroids were reduced to a maintenance dose (0.5 to 1 g/kg/day of oral prednisolone). The daily dose of oral prednisolone was decreased by 5–10 mg/week, depending on the patient’s response.
The cytokine interleukin-6 (IL-6) plays a critical role in the inflammatory process and is implicated in the development of severe acute respiratory syndrome[36]. A preclinical investigation demonstrated that EGFR-TKI administration not only decreased the viability of cancer cells, but also increased IL-6 production from cancer cells[37]. The results indicated that EGFR-TKIs could induce ILD via IL-6 production[37]. Therefore, blocking IL-6 activity should be effective for controlling EGFR-TKI-induced ILD. The recombinant humanized anti-human IL-6 receptor monoclonal antibody tocilizumab is widely used for treating autoimmune diseases such as giant cell arthritis[38]. Currently, tocilizumab can achieve a clinical response in patients with severe respiratory failure caused by severe acute respiratory syndrome coronavirus 2 infection[39][40]. No clinical evidence was established in the area of drug-induced ILD; however, a similar pathogenesis should exist. Thus, tocilizumab may be effective in treating severe cases of EGFR-TKI-induced ILD.
In addition to the previously approved use of corticosteroids as the sole therapy for IPF, pirfenidone and nintedanib have been approved recently and used clinically for the treatment of IPF. In particular, nintedanib has been shown to prevent the development of acute IPF exacerbations[41]. The indications for these anti-fibrotic drugs are currently limited to IPF, although the preventive effect of these drugs with regard to acute exacerbation, due to surgery and chemotherapy, in lung cancer patients with pre-existing IPF is being investigated[42][43]. Kanayama et al. classified 100 patients with indications for surgery into three groups according to the surgical risk score and conducted a retrospective study on the incidence of IPF as an acute exacerbation in the pirfenidone-administered and control groups[43]. The results showed that the incidence of acute exacerbation was suppressed by pirfenidone in all groups. A randomized controlled study is currently underway to evaluate the effect of nintedanib for preventing acute IPF exacerbation during CBDCA + Nab-paclitaxel combination therapy for NSCLC patients with IPF[44]. Thus far, only case studies have reported the effects of anti-fibrotic drugs on EGFR-TKI-related ILD; however, a clinical study of this side effect of EGFR-TKI-related ILD has been planned and will finally help to develop a safer treatment regimen for EGFR-TKI-based therapy