Time to treatment discontinuation in first-line non-small cell lung carcinoma: an overview

Abstract

Objective

Time To Discontinuation (TTD) is defined as the time from the start of treatment to the end of treatment, usually occurring due to loss of efficacy or occurrence of adverse events. It has become an important surrogate efficacy endpoint especially in real-world studies due to its correlation with endpoints such as Progression Free Survival (PFS). The aim of the study is to conduct a literature review of all studies reporting TTD in first-line therapy of Non-Small Cell Lung Cancer (NSCLC).

Methods

All articles that reported TTD for any first-line treatment of NSCLC as of 30 June 2022 were extracted from the PubMed search engine. From these articles, the drugs, study type, and TTD values were extracted. A descriptive analysis of the studies was made, dividing the TTD by subgroup according to the type of treatment (traditional chemotherapy, target therapy, immunotherapy) and study design (clinical trials, real world studies).

Results

Fifty-five studies were considered for the analysis, of which 12 were published in 2021; 28 were clinical trials and 27 were real-world studies. Thirty of the studies considered involved conventional chemotherapy and expressed TTD values from 1.4 to 4.5 months, 5 of the studies considered involved immunotherapy with TTD values from 2.1 to 7.4 months and 18 of the studies considered target therapy, with TTD values from 4 to 31 months. The clinical trials reported TTD values from 1.4 to 16 months and the real-world studies from 2 to 31 months.

Conclusion

Studies reporting TTD are increasing, most notably real-world studies. Given the increasing importance of TTD as an efficacy endpoint, it becomes critical to measure and monitor it in various therapeutic settings such as NSCLC. This is the first study to review all TTD values of drugs used in first-line NSCLC.

Keywords:

• Time to treatment discontinuation
• non small cell lung cancer
• first line
• real world studies

Introduction

Non-Small Cell Lung Cancer (NSCLC) is a type of cancer affecting the epithelial tissues lining the bronchi and lung parenchyma, caused by pilot oncogenic mutations that have been identified mainly in adenocarcinomas.¹ In recent years, the range of therapies available for the treatment of NSCLC has expanded, with molecular-targeted therapies (target therapy) and immunotherapeutic drugs being added to traditional chemotherapy.^2–4 In order to be marketed for NSCLC, these drugs had to prove their efficacy in pre-marketing randomized clinical trials in terms of survival time, as measured by robust primary outcomes such as overall survival (OS) and progression free survival (PFS).⁵

As is now well known, during the post-marketing experience these drugs are used in conditions that are not identical to those of clinical trials and on patients who would sometimes have been considered in trials as not eligible for treatment.⁶ Thus, in recent years post-marketing studies, or real-world studies (RWS), have become increasingly important in providing additional efficacy and safety data that can support and complement the evidence gathered in clinical trials in order to improve the benefit-risk profile of marketed drugs.⁷

Time To Treatment Discontinuation (TTD) is a surrogate efficacy endpoint used mainly in RWS, which is faster and more convenient than other endpoints and has been found to correlate with PFS in both RCTs and RWSs.⁸

TTD is defined as the time elapsed between the first administration of the drug and its discontinuation, which may occur due to disease progression or because of the occurrence of serious adverse events attributable to the therapy and may also be identified by other synonyms such as persistence, treatment duration, time on treatment.⁹

It is therefore important to investigate this new efficacy endpoint to understand its prevalence and to obtain a current status report. This study is an overview aimed at analysing the treatment duration data reported in the literature for first-line treatment of NSCLC.cos

Materials and methods

The literature search was conducted on 30 April 2022 using the PubMed search engine, the search string used was (first-line OR naive OR untreated) AND ((NSCLC) OR (non-small-cell-long-cancer)) AND ((time-to-discontinuation) OR (TTD) OR (time-to-treatment-discontinuation) OR (persistence) OR (time-on-treatment) OR (drug-survival) OR (treatment-duration)).

All studies reporting the duration of treatment with first-line anticancer drugs in NSCLC were included, regardless of how duration was defined, and irrespective of the design and the drug used.

On the other hand, studies that did not report TTD for first-line treatment with NSCLC indication, dose-finding studies, reviews, expert opinions, unpublished English-language papers, non-antitumour drugs, treatments with a maximum number of cycles regardless of clinical response, and evaluations of treatment sequences were excluded.

For each study, the following data were extracted: sample size of patients treated, study design, type of NSCLC, drug administered, inclusion and exclusion criteria, method of calculating TTD, duration and whether this was a study endpoint. For studies with infusion therapies in which the duration of treatment was not reported but the median number of cycles performed, the duration of treatment was calculated by multiplying the median number of cycles by the number of days in each cycle.

The results were presented in a descriptive manner reporting the characteristics of the studies, the sample of patients and treatment under study, as well as the method of calculating TTD. In addition, the TTD endpoint was reported by subgroup according to the type of treatment (dividing between traditional chemotherapy, target therapy and immunotherapy) and study design. The boxplots were created using BoxPlotR (http://shiny.chemgrid.org/boxplotr/).

Results

A total of 724 articles were extracted from the literature search, of which 55 studies were considered by applying the exclusion criteria, as shown in the flowchart in Figure 1.

Figure 1. Flowchart of the studies considered for the overview.

The results are summarized in Table 1.

Table 1. study characteristics and endpoint.

Author	Type of Study	TTD endpoint	Treatment	N	mTTD (months)	Method of calculation of TTD
Agulnik JS, 2021	RWS	Primary	Afatinib	4	11.4	Duration of treatment was calculated as the time (in months) elapsed between the start and end dates of the treatment
Agulnik JS, 2021	RWS	Primary	Erlotinib	56	11.5	Duration of treatment was calculated as the time (in months) elapsed between the start and end dates of the treatment
Agulnik JS, 2021	RWS	Primary	Gefitinib	110	11.7	Duration of treatment was calculated as the time (in months) elapsed between the start and end dates of the treatment
Agulnik JS, 2021	RWS	Primary	Targeted therapies	170	Not reached	Duration of treatment was calculated as the time (in months) elapsed between the start and end dates of the treatment
Ariyasu R, 2018	RWS	–	Osimertinib with a T790M ratio of ≥0.4	13	31	Not specified
Ariyasu R, 2018	RWS	–	Osimertinib with a T790M ratio of <0.4	20	18.9	Not specified
Belani CP, 1999	CT	–	Docetaxel/Cisplatin	47	3.5	Number of cycles
Belani CP, 2000	CT	–	Docetaxel/Carboplatin	33	3.5	Number of cycles
Beretta GD, 2000	CT	–	Gemcitabine/vinorelbine	43	2.8	Number of cycles
Dansin E, 2012	CT	–	bevacizumab with chemotherapy	2212	4.9	Number of cycles
Davis KL, 2018	RWS	Co–primary	Crizotinib	137	9	From initiation to last observed dose
Dranitsaris G, 2005	RWS	Primary	Gemcitabine	143	3.15
Dranitsaris G, 2005	RWS	Primary	Vinorelbina	1361	3.7
Fukuoka M, 1992	CT	–	Irinotecan	73	1.4	Number of cycles
Fukuoka M, 1992	CT	–	Mitomicina/VIndesina/Cisplatino	49	2.3	Number of cycles
Galetta D, 2002	CT	–	Gemcitabine and docetaxel	47	3.7	Number of cycles
Gan CL, 2021	RWS	Primary	Pembrolizumab Elegible	157	5.9	Treatment discontinuation (TD) was defined as the time from the initiation of 1 L IO treatment to treatment discontinuation for any reason
Gan CL, 2021	RWS	Primary	Pembrolizumab Non Elegible	242	2.1	Treatment discontinuation (TD) was defined as the time from the initiation of 1 L IO treatment to treatment discontinuation for any reason
Ganti AK, 2022	RWS	Primary	Alectinib	445	8.8	as switching to another ALK inhibitor or a gap of at least 60 days without treatment with the same ALK inhibitor (whichever came rst)
Ganti AK, 2022	RWS	Primary	Crizotinib	1037	27.1	as switching to another ALK inhibitor or a gap of at least 60 days without treatment with the same ALK inhibitor (whichever came first)
Georgoulias V, 2003	CT	–	Docetaxel/Cisplatin	132	2.1	Number of cycles
Georgoulias V, 2003	CT	–	Docetaxel/Gemcitabine	114	3	Number of cycles
Georgoulias V, 1998	CT	–	Vinorelbine + Cisplatin  + Docetaxel	24	2.8	Number of cycles
Hata A, 2015	CT	–	Pemetrexed	26	2.1	number of cycles
Hellmann, 2020	CT	Secondary	Chemoterapy	570	2.6	–
Hellmann 2020	CT	Secondary	Nivolumab + ipilimumab	576	4.2	–
Herbst RS, 2005	CT	–	Erlotinib/Carboplatin	526	4.6	–
Hochmair MJ, 2018	RWS	Primary	Afatinib	204	11.9	Time from the first dose of afatinib to that of the last dose of osimertinib or death
Horvat P, 2021	RWS	Primary	Nivolumab, pembrolizumab, or atezolizumab	2035	4.9	The length of time from the index date that the patient initiates treatment with a PD-(L)1 inhibitor to the date that the patient discontinues the treatment
Ito K, 2021	RWS	Primary	Afatinib	224	18.6	The period from the induction of afatinib/osimertinib in first- line settings to the date of discontinuation of afatinib/ osimertinib if absence of follow up osimertinib/afatinib therapy, or the period from the same point to the discontinuation of sequential osimertinib/afatinib therapy if the patient received follow up osimertinib/afatinib therapy.
Ito K, 2021	RWS	Primary	Osimertinib	326	20.5	The period from the induction of afatinib/osimertinib in first- line settings to the date of discontinuation of afatinib/ osimertinib if absence of follow up osimertinib/afatinib therapy, or the period from the same point to the discontinuation of sequential osimertinib/afatinib therapy if the patient received follow up osimertinib/afatinib therapy.
Jahanzeb M, 2004	CT	–	Docetaxel/Carboplatin	25	4.2	Number of cycles
Jahanzeb M, 2020	RWS	Primary	Crizotinib, alectinib or ceritinib	n = 492 received crizotinib, alectinib (n = 80) or ceritinib (n = 9)	7.04	Time from initiation of current ALK TKI line to the end of the current line if there were any of the following: (a) initiation of next-line ALK TKI, (b) gap of more than 120 days between current ALK TKI therapy and last follow-up date, or (c) death; otherwise, it was censored at the last follow-up.
John A, 2021	RWS	Primary	Any therapy – adherent patients	11112	2.3	Duration between the first and the last observed drug episode in the first-line setting
John A, 2021	RWS	Primary	any therapy – nonadherent patients	4785	2.4	Duration between the first and the last observed drug episode in the first-line setting
John A, 2021	RWS	Primary	Chemotherapy adherent patients	5005	3.45	Duration between the first and the last observed drug episode in the first-line setting
John A, 2021	RWS	Primary	Immunotherapy adherent patients	1196	4.57	Duration between the first and the last observed drug episode in the first-line setting
John A, 2021	RWS	Primary	Targeted therapy adherent patients	4906	9.4	Duration between the first and the last observed drug episode in the first-line setting
Kallab AM, 2005	CT	–	Paclitaxel weekly/Carboplatin	29	3.3	Number of cycles
Kourousis C, 1998	CT	–	Vinorelbine + Docetaxel	46	2.1	Number of cycles
Kwon MR, 2002	CT	–	Vinorelbine/Mitomycin C/Cisplatin	30	3.7	Number of cycles
Lee C, 2021	RWS	–	Osimertinib	52	14.2	Not specified
Lee J, 2021	RWS	Secondary	Cytotoxic chemotherapy was the most common approach in first–line therapy (69.9%), followed by targeted therapy (13.0%), immunotherapy (6.5%), and chemotherapy plus immunotherapy combination therapy (2.8%)	216	4.5	The interval between initiation and termination of a therapy regimen
Lee SY, 2021	RWS	Primary	Afatinib	419	17.8	Time from the first dose of afatinib to the date of treatment discontinuation or death
Lester J, 2021	RWS	Primary	Any therapy	1003	2.1	Interval between 1 L treatment initiation and discontinuation for any reason, including death
Lester J, 2021	RWS	Primary	Chemotherapy	698	2.1	Interval between 1 L treatment initiation and discontinuation for any reason, including death
Lester J, 2021	RWS	Primary	Immunoherapy	179	5.3	Interval between 1 L treatment initiation and discontinuation for any reason, including death
Lester J, 2021	RWS	Primary	Targeted therapies	126	7.6	Interval between 1 L treatment initiation and discontinuation for any reason, including death
Lim J, 2019	RWS	Primary	Afatinib	550	9.9	Time from the index date to the end of treatment. End of treatment was defined as either date of treatment discontinuation (>60-day gap in supply of the index drug from the last day of supply) or the addition of a chemotherapy treatment before discontinuation of the index medication, whichever came first
Lim J, 2019	RWS	Primary	Erlotinib	2602	12	Time from the index date to the end of treatment. End of treatment was defined as either date of treatment discontinuation (>60-day gap in supply of the index drug from the last day of supply) or the addition of a chemotherapy treatment before discontinuation of the index medication, whichever came first
Lorenzi M, 2022	RWS	Primary	Osimertinib	126	25.3	measured from the osimertinib start to discontinuation for any cause
Lynch TJ Jr, 1994	CT	–	Topotecan	20	2.45	Not specified
Monnerat C, 2004	CT	–	Pemetrexed/Gemcitabine	60	2.8	Number of cycles
Mori K, 1999	CT	–	Irinotecan and cisplatin	41	2.2	Number of cycles
Mudad R, 2017	RWS	Secondary	Gemcitabine plus carboplatin	132	3.5	TTD was defined as the time between the first and last dose of nab-paclitaxel or gemcitabine plus 7 days. Treatment discontinuation was defined as the initiation of a new regimen (different from first line) or >60-day interruption of either nab-paclitaxe
Mudad R, 2017	RWS	Secondary	Nab–paclitaxel plus carboplatin	61	4.3	TTD was defined as the time between the first and last dose of nab-paclitaxel or gemcitabine plus 7 days. Treatment discontinuation was defined as the initiation of a new regimen (different from first line) or >60-day interruption of either nab-paclitaxel or gemcitabine. Patients were censored if the last date of administration was <30 days before the data cutoff
Nadler E, 2021	RWS	Primary	Systemic chemotherapies	5859	2	The interval between 1 L treatment initiation and discontinuation for any reason (including death)
Nadler E, 2021	RWS	Primary	Immuno Oncology combination regimens	324	2.8	Kaplan – Meier method
Nadler E, 2021	RWS	Primary	Immuno Oncology monotherapies	907	3	Kaplan – Meier method
Nadler E, 2021	RWS	Primary	Targeted therapies	656	3.5	Kaplan – Meier method
Nomura F, 1987	CT	–	Mytomicin/Vinblastine/Cisplatin	30	2.8	Number of cycles
Ohnoshi T, 1991	CT	–	Ifosfamide/Cisplatin/Vindesine	27	2.8	Number of cycles
Planchard D, 2019	CT	Secondary	Gefitinib or Erlotinib	277	11.5	TDTKI was defined as the time from randomization to the last dose of any EGFR-TKI, or death, without interruption by any non-EGFR–TKI therapy
Planchard D, 2019	CT	Secondary	Gefitinib or erlotinib	277	16	TDTKI was defined as the time from randomization to the last dose of any EGFR-TKI, or death, without interruption by any non-EGFR–TKI therapy
Planchard D, 2019	CT	Secondary	Osimertinib	279	20.8	TDTKI was defined as the time from randomization to the last dose of any EGFR-TKI, or death, without interruption by any non-EGFR–TKI therapy
Planchard D, 2019	CT	Secondary	Osimertinib	279	23	TDTKI was defined as the time from randomization to the last dose of any EGFR-TKI, or death, without interruption by any non-EGFR–TKI therapy
Saito H, 2006	CT	–	Irinotecan 3–week scheduling/Cisplatin	27	2.8	Number of cycles
Saito H, 1990	CT	–	Ifosfamide, cisplatin, vinblastine	42	1.87	Not specified
Sørensen JB, 1999	CT	–	Gemcitabine and vindesine	42	3.3	Number of cycles
Spigel DR, 2005	CT	–	Gefitinib	70	4	Not specified
Takada M, 1998	CT	–	Gemcitabine	73	1.87	Number of cycles
Trédaniel J, 2009	CT	–	Gemcitabine/vinorelbine/ oxaliplatin	39	4	Not specified
Velcheti V, 2022	RWS	Primary	Pembrolizumab ECOG 0 – 1	807	2.1	[(date of last recorded pembrolizumab dose – date of first recorded dose) + 1 day], defining treatment discontinuation at the last noncancelled order or administration date if patients died during pembrolizumab therapy or initiated a next line of therapy, or if there was a gap of 120 days between the last recorded pembrolizumab dose and last clinical contact date in the dataset
Velcheti V, 2022	RWS	Primary	Pembrolizumab ECOG 2	237	7	[(date of last recorded pembrolizumab dose – date of first recorded dose) + 1 day], defining treatment discontinuation at the last noncancelled order or administration date if patients died during pembrolizumab therapy or initiated a next line of therapy, or if there was a gap of 120 days between the last recorded pembrolizumab dose and last clinical contact date in the dataset
Velcheti V, 2022	RWS	Primary	Pembrolizumab KRAS mutated	164	7.4	[(date of last recorded pembrolizumab dose – date of first recorded dose) + 1 day], defining treatment discontinuation at the last noncancelled order or administration date if patients died during pembrolizumab therapy or initiated a next line of therapy, or if there was a gap of 120 days between the last recorded pembrolizumab dose and last clinical contact date in the dataset
Velcheti V, 2022	RWS	Primary	Pembrolizumab KRAS unknown	214	7.6	[(date of last recorded pembrolizumab dose – date of first recorded dose) + 1 day], defining treatment discontinuation at the last noncancelled order or administration date if patients died during pembrolizumab therapy or initiated a next line of therapy, or if there was a gap of 120 days between the last recorded pembrolizumab dose and last clinical contact date in the dataset
Velcheti V, 2022	RWS	Primary	Pembrolizumab KRAS wild type and ECOG 0 – 1	166	7.9	[(date of last recorded pembrolizumab dose – date of first recorded dose) + 1 day], defining treatment discontinuation at the last noncancelled order or administration date if patients died during pembrolizumab therapy or initiated a next line of therapy, or if there was a gap of 120 days between the last recorded pembrolizumab dose and last clinical contact date in the dataset
Waits TM, 1992	CT	–	Etoposide	25	1.4	Number of cycles
Weiss J, 2017	RWS	Primary	3 w Paclitaxel	153	2.6	The time between the first and last taxane doses + 21 days
Weiss J, 2017	RWS	Primary	Nab Paclitaxel	114	2.8	The time between the first and last taxane doses + 7 days
Weiss J, 2017	RWS	Primary	Weekly Paclitaxel	208	3.8	The time between the first and last taxane doses + 7 days
Wu S, 2020	RWS	Secondary	Afatinib	91	12.1	Not specified
Wu S, 2020	RWS	Secondary	Erlotinib	154	14.4	Not specified
Wu S, 2020	RWS	Secondary	Gefitinib	162	16.2	Not specified
Yang JC et al., 2020	RWS	Primary	Afatinib	272	4.2	Time to treatment failure, defined as time from the start of therapy to treatment discontinuation for any reason, or death
Yang JC et al., 2020	RWS	Primary	Afatinib	–	4.5	Time to treatment failure, defined as time from the start of therapy to treatment discontinuation for any reason, or death
Yang JC et al., 2020	RWS	Primary	Afatinib	–	10.8	Time to treatment failure, defined as time from the start of therapy to treatment discontinuation for any reason, or death
Yang JC et al., 2020	RWS	Primary	Afatinib	–	14.7	Time to treatment failure, defined as time from the start of therapy to treatment discontinuation for any reason, or death
Zhang Q., 2015	RWS	–	Pemetrexed/cisplatin	41	3.7	Number of cycles

Abbreviations. CT, Clinical Trial; RWS, Real World Studies.

Over the years, there has been an increase in studies reporting TTD, as shown in Figure 2. In the past it was mainly evaluated in CTs, whereas in recent years RWSs report it more frequently.

Figure 2. Number of articles reporting treatment duration, by year of publication and divided between RWSs and CTs.

The duration of treatment was defined differently in the various studies: Whilst in CT it is almost never reported as a primary or secondary endpoint but rather among the information on the use of the drug or sample characteristics, in pharmacy utilization and effectiveness studies in Real World studies it very often represents a primary or secondary efficacy endpoint.

In most cases, “treatment duration” was defined in 43 out of 55 studies either as the time between the first and last drug intake or discontinuation^10–19 without giving further details in the calculation methodology, otherwise as the time between the first and last drug administration + 1 day,²⁰ 7 days,²¹ 21 days.²²

There were 28 Clinical trials (CTs) and 27 RWSs so they were equally represented.

The boxplot in Figure 3 shows the distribution of TTD in CTs and RWSs, which were 1.4 to 16 months for CTs and 2 to 31 months for RWSs.

Figure 3. TTD values in months for CTs and RWSs.

Looking at the individual drugs, it was noted that the highest mTTD (median TTD) were related to target therapies, specifically Afatinib and Osimertinib, which were the most widely reported molecules, and which were almost always related to RWSs.

In terms of the type of therapy considered, 30 studies considered traditional chemotherapies, with Gemcitabine being the most cited; 18 studies considered Target Therapies, with Afatinib being the most widely cited molecule; immunotherapies were considered in 5 studies, of which 3 with Pembrolizumab and the remaining 2 studies considered all the treatments used in the first-line treatment of NSCLC.

The boxplot in Figure 4, on the other hand, shows the distribution by drug class, with traditional chemotherapies showing TTD values from 1.4 to 16 months, immunotherapies showing TTD values from 2.1 to 7.4 months and target therapies showing TTD values from 4 to 31 months.

Figure 4. TTD values in months by drug class.

Dividing the studies into subgroups, the drugs considered within CTs differ from those considered within RWSs. In fact, traditional chemotherapeutics were more widely reported in CTs (24 out of 28 studies), whilst target therapies were more widely reported in RWSs (14 out of 27). Afatinib, Alectinib and Osimertinib were only to be found in RWSs and not in CTs.

The sample size was extremely variable, with a median of 73 patients and a range of 4 to 11112 patients; the sample sizes for CTs and RWSs were different, with a median value of 42 for CTs and 216 for RWSs.

Discussion

Target therapy and immunotherapy show higher TTD than conventional chemotherapy, confirming the efficacy data evaluated with more robust efficacy endpoints.

Target therapies have higher mTTD values than both immunotherapies and conventional chemotherapy. It must be specified that the target therapies that showed the highest TTD (Osimertinib, Afatinib and Alectinib) were evaluated exclusively in RWSs and never in CTs. Immunotherapies showed higher TTD than conventional chemotherapy and lower TTD than target therapies, albeit in fewer studies than the other therapies (5 studies), all of which were carried out under Real World conditions.

CTs and RWSs are different kinds of studies and have different objectives. Nevertheless, we decided to consider them both since the aim of this review is to provide an overview of the literature reporting on treatment duration in first-line NSCLC, irrespective of study type and treatment.

CTs are the gold standard for evaluating the efficacy of a treatment, whilst RWSs can have different objectives and consider several treatments at once, thus providing an insight of how drugs are actually used in clinical practice.

Figure 2 shows how the number of RWSs reporting TTD has exploded in recent years, in line with the role of real-world evidence and an increasing number of publications in this area. TTD has in fact taken on the important value of a surrogate efficacy endpoint in RWSs in recent years, and for this reason has often been evaluated as a primary outcome in RWSs on drug utilization.

Prior to this “revolution”, the duration of therapy was simply part of the description of how the drug was used in clinical trials or seen as part of the characteristics of the patient sample, where it was almost always not even considered among the endpoints.

There are 2 critical issues in the definition and calculation of treatment duration:

(1) The definition of treatment discontinuation: after how many days since the last dispensing/administration of medication can a treatment be deemed “discontinued”, and not simply suspended for a period? The value is at the discretion of the authors and varies from 60^21,23,24 to 120 days^20,.²⁵ Therefore, depending on the choice of interval, the number of patients considered to be persistent on treatment may vary significantly. For this reason, to assess the statistical robustness of the results, some studies such as Ganti et al.²³ also perform a sensitivity analysis using a gap <30 days in the definition of persistence.

The choice of the maximum interval is an important aspect in the calculation of persistence and should be made according to the drug analysed and according to the dosage and the quantity dispensed.

Compliance with these variables would help standardize the method of analysis, making the results comparable. Certainly, the details of the methodology are necessary to better interpret the results presented and to lay the foundations for comparisons with other similar studies.

(2) The method by which the duration of treatment is calculated is also not unequivocal. For example, in the case of intravenous drugs, there are studies that add up the duration of an entire cycle to the time between the first and last administration,²² studies that add up to only 1 day²⁰ and studies that do not describe the calculation method at all.

In the case of oral drugs, no study specifically describes the method of calculating the duration of treatment, so it is not clear whether the date of last administration was determined by observation, questionnaire or by adding the number of days of dispensed dosage units to the last date of dispensing of the drug, or whether this corresponds to the last date of dispensing of the drug.

As far as the sample size of patients is concerned, it differs significantly between CTs and RWSs. In fact, the sample size in CTs is predetermined on the basis of statistical methods and expected results,²⁶ whereas in RWSs the sample size cannot be established a priori but reflects the availability of patients taking the treatment in clinical practice.

Conclusion

Given the great interest and large production of RWSs in recent years, the availability of data on treatment duration will continue to increase. In the first-line treatment of NSCLC it would be interesting to compare the values in RWSs with those in CTs, to assess whether in clinical practice the number and timing of discontinuations are comparable to those in CTs.

Transparency

Declaration of funding

No funding was received to assist with the preparation of this manuscript.

Declaration of financial/other relationships

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

RL contributed to the conception and design of the study, contributed to the acquisition, analysis, interpretation of data, and gave the final approval and agrees to be accountable for all aspects of the work in ensuring that questions relating to the accuracy or integrity of any part of the work are appropriately investigated and resolved. AZ, VI, AR, FM and FS contributed to acquisition, analysis, interpretation of data, and gave the final approval.

Acknowledgements

None.

Data availability statement

All data generated or analysed during this study are included in this published article.