This study included randomized controlled trials (RCTs), prospective open-label trials and single-arm prospective studies The language in which it was written was restricted to English. The search was limited to studies published from 2005 through 2025, as the first scientific evidence supporting the efficacy of epidermal growth factor receptor tyrosine kinase inhibitors in the treatment of non-small cell lung cancer emerged around 2003. Since then, both their clinical use and the body of evidence have evolved substantially. Accordingly, a 20-year time frame was considered appropriate and sufficiently comprehensive for the purposes of this study. The studies were included if they had patients with NSCLC who were about to initiate treatment with EGFR TKIs (erlotinib, gefitinib, afatinib, dacomitinib, osimertinib) and were starting at the same time, a preventive treatment with oral tetracyclines due to the appearance of acneiform rash. Studies in which the primary or secondary endpoint was the incidence of acneiform rash were included, whereas studies evaluating exclusively topical preventive treatments were excluded. The primary outcome assessed was the incidence of acneiform rash of any grade.

A search across the scientific databases PubMed, Cochrane and Web of Science was conducted on January 2025, by 2 authors, using the following terms: “(Prophylactic Treatment OR Preventive treatment OR Pre-emptive treatment) AND Epidermal Growth Factor Receptor-Tyrosine Kinase Inhibitor AND (Skin Toxicities OR Acneiform eruption OR Acneiform Rash) AND Non-small Cell Lung Cancer”. Additionally, citations from relevant articles were read as well. A screening phase was conducted by both authors, reading the title and abstract of all extracted articles from the search. From there, all screened articles were assessed for eligibility and were fully read. Those that met the inclusion criteria were included in the review.

Data extracted from each study included the incidence of acneiform rash of any grade and of grade 2 or higher, for both the control and intervention arms, which were subsequently compared. Three included studies lacked a control arm; therefore, the incidence of acneiform rash of any grade in these studies was analyzed and compared with rates reported in major clinical trials, including ARCHER 105017 and LUX-Lung 8.5 Acneiform rash of any grade was defined as grade 0–5 rash and encompassed various reported terms, including acneiform rash,18 rash/acne,19 skin rash,20 rash and dermatitis acneiform,21 rash/folliculitis,22 and rash.23,24 In addition, data were collected on the proportion of patients who required dose reduction or treatment discontinuation of epidermal growth factor receptor tyrosine kinase inhibitor therapy, as well as intervention characteristics, including oral tetracycline dosage, duration of prophylactic treatment, type of epidermal growth factor receptor tyrosine kinase inhibitor used, and its dosage.

To assess the quality of included trials 2 different tools were used. The version 2 of the cochrane risk-of-bias tool for randomized trials (RoB 2) was applied to assess the risk of bias of the RCTs and the prospective open-label trials with both control and experimental groups and version 2 of The Risk of Bias Assessment Tool for Nonrandomized Studies (RoBANS 2) for the non-randomized, prospective single-arm trials.

The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) process was followed and the exclusion of the studies at each stage are shown in the flowchart (Fig. 1). A total of 55 articles were retrieved from this search and 5 more after reading the citations of relevant articles. After reading title and abstract, a total of 47 articles were excluded, 9 articles were fully read and assessed for eligibility and finally 7 were included in the review. The excluded trials25,26 appeared to meet all the inclusion criteria; however, they included both NSCLC and GI cancer patients, and treatment with both EGFR TKIs and anti-EGFR monoclonal antibodies, such as cetuximab, without reporting cancer type-specific results, thereby precluding further analysis.

Fig. 1.

PRISMA flow-chart.

We included a total of 7 trials in our systematic review. All the trials tested for the preventive treatment with oral tetracyclines—4 with minocycline, 2 with doxycycline and 1 with tetracycline—in patients with NSCLC on EGFR TKIS, such as erlotinib, afatinib and dacomitinib. Table 1 illustrates the characteristics of the included studies.

The results of the risk of bias assessment using version 2 of the cochrane risk-of-bias tool for randomized trials (RoB 2) and version 2 of The Risk of Bias Assessment Tool for Nonrandomized Studies (RoBANS 2) are shown in Figs. 2 and 3, respectively.

Fig. 2.

Risk of bias using RoB2.

Fig. 3.

Risk of bias using RoBANS2.

The studies assessed through RoB2 had an overall low risk of bias. Deplanque et al.,22 Arrieta el al.,23 and Melosky et al.,24 are open-label trials with a higher risk of observer bias. The single-arm studies, assessed through RoBANS2, have inevitably a higher risk of confounding due to the lack of a comparable control group. Furthermore, they have a higher risk of observer bias due to their open label nature.

In 4 of the included trials, the incidence rate of all-grade rash was evaluated and compared between the experimental group on preventive therapy with oral tetracyclines for acneiform rash and the control group, without any form of preventive therapy. The prophylactic intervention was initiated at the start of the EGFR TKI therapy in all trials. The duration of prophylactic therapy until the evaluation of skin toxicities varied across trials (from 4 to 16 weeks), and it is showed individually for each trial in Table 1.

Doxycycline in the dosage of 100mg twice daily was effective preventing the acneiform rash in Lacouture et al.21 in patients on dacomitinib, reducing the incidence rate of all-grade rash from 74.2% in the control group to 57.2% in the doxycycline group, and with a relative risk of rash and dermatitis acneiform of 0.62 and 0.39, respectively. The G≥2 rash incidence rate reported in the control group was 31% and in the doxycycline group, 16.1%. In the study by Deplanque et al.,22 among patients receiving erlotinib with prophylactic doxycycline 100mg daily, the difference in the incidence of all-grade rash between the control group and the doxycycline group was not statistically significant (81% vs 71%; p=.18). Doxycycline decreased the rate of severe rash, with an incidence rate of grade 3 (G3) rash of 19% in the control arm and 4% in the doxycycline arm (p<0.001). These results are shown in Table 2.

Tetracycline, in the dosage of 250mg administered twice daily reduced the incidence rate of any grade rash (75.6% vs 44.5%; p=0.046) and G≥2 rash (35.6% vs 15.6%; p=0.030) in Arrieta et al.23 in patients on afatinib. These results are shown in Table 3.

For minocycline in the dosage of 100mg twice daily, Melosky et al.24 did not find a reduction in the incidence rate of all-grade rash in patients on erlotinib between the prophylactic treatment arm and the control arm (82 vs 84%; p=0.8769). However, the incidence rate of G3 rash was significantly different between the control arm and the prophylactic treatment arm (28% and 12%, respectively; p=0.0455).

Minocycline, 100mg daily was the prophylactic treatment of the single-arm prospective studies included in this review. Prophylactic treatment was initiated at the same time as the EGFR TKI therapy in all these trials. Okajima et al.19 and Ichiki et al.20 assessed prophylactic treatment with minocycline in patients on afatinib. They found incidence rates of all-grade rash of 50% and 44.80%, respectively, and an incidence rate of G≥2 rash of 20% and 3.4%, respectively. The incidence rate of all-grade rash in the Lux-Lung 85 trial was 67% and the incidence rate of G≥2 was 6%. This was a trial with patients with NSCLC on afatinib in whom no prophylactic measures for skin adverse effects were taken.

Iwasaku et al.18 tested prophylactic treatment with 100mg of minocycline in patients on dacomitinib. The incidence rate of all-grade rash was 68,3% (26.8% for G≥2). In the ARCHER 105017 trial, the incidence rate of all-grade rash in patients on dacomitinib and without any form of prophylactic treatment for dermatologic adverse effects was 77.7% (25.3% for G≥2). The results regarding minocycline are shown in Table 4.

Furthermore, dose reduction and treatment discontinuation were analyzed across the studies and the data is shown in Tables 5 and 6. The available data were somewhat heterogeneous. In 3 randomized clinical trials,22,23 information on dose reductions was reported for both the control and intervention groups, without specification of the underlying cause. Dose reduction was higher in the control group vs the experimental group in Deplanque et al.22 [43% vs 25% (p=0.02)]. In Arrieta et al.,23 however, dose reduction was lower in the control group vs the experimental group [46.6% vs 53.4% (p=0.378)], as well as in Lacouture et al.21 (24.2% vs 28.8%). In the study by Melosky et al.,24 the percentages of dose reduction and treatment discontinuation were not reported. In the single-arm studies, we found that in Iwasaku et al.,18 dose reduction occurred in 19.5% of patients due to skin toxicities, and in 14.6% of these due to acneiform rash. In the study by Okajima et al.,19 dose reduction occurred in 58.7% of patients for all causes, with 13% attributable to acneiform rash. In the study by Ichiki et al.,20 an overall dose reduction rate of 62% was reported, without further specification.

Data of treatment discontinuation was available in Iwasaku et al.,18 occurring in 22.2% of patients due to disease progression; in Okajima et al.,19 occurring in 13% of patients due to G4 transaminase elevation, G3 ileitis, G2 paronychia, G2 decrease appetite and G2 diarrhoea, and in Lacouture et al.21 occurring in 22.7% of patients from the control group and 18.2% of patients from the experimental group, without specification of the cause.

This systematic review included a total of 7 trials, all testing for the prevention of skin toxicities with oral tetracyclines in patients with NSCLC on EGFR TKIs.

Among the 4 trials comparing a control arm with an oral tetracycline arm, 2 demonstrated a significant reduction in the incidence of all-grade rash with oral tetracyclines. In the remaining 2 trials, the difference in all-grade rash incidence between groups was not statistically significant; however, prophylactic treatment was associated with a reduced incidence of severe rash. All comparative trials reported oral tetracyclines to be well tolerated. In the single-arm studies, the overall incidence of all-grade rash was lower than that reported in the major comparator trials, ARCHER 105017 and LUX-Lung 8.5 Oral tetracyclines were also well tolerated in these studies.

Regarding dose reduction and treatment discontinuation, the heterogeneity of the results makes it difficult to analyze any possible patterns.

We can state that there is a significant percentage of patients who undergo dose reduction when on EGFR TKIs and an important part is due to skin toxicities, such as acneiform rash. Thus, dose reduction is a real issue with this therapy. Regarding the impact of tetracyclines in dose reduction, one trial found that the group exposed to oral tetracyclines had less dose reductions and 2 found the control group to have less dose reductions, so we cannot securely state that oral tetracyclines reduce the percentage of dose reduction in these patients.

Treatment discontinuation is also an important issue. It occurs in a significant percentage of patients. In our review, 1 trial found that in patients on prophylactic treatment with oral tetracyclines there was a smaller percentage of treatment discontinuation vs the control group.

When a patient is starting treatment with an EGFR TKI, the possibility of developing a rash, and even a severe rash is >50% and 15% respectively. Reducing the chances of this event should be a priority for the physician since these adverse effects have such an impact on patients’ lives and can strongly affect treatment adherence. Taking into consideration that oral tetracyclines are well tolerated by patients, starting an oral tetracycline concomitantly with EGFR TKI treatment should be considered by their physicians.