Chemotherapy remains an essential element of personalized care for persons with lung cancers

Chemotherapy works. Chemotherapy can enhance the effectiveness of all other modalities and improve the curability of patients with locoregional spread of their lung cancers. While tremendous advances in molecularly targeted and immunotherapies are quickly changing the treatment landscape, chemotherapy remains an essential component of personalized care for persons with lung cancers.

Abstract

Molecularly targeted and immunotherapies have improved the care of patients with lung cancers. These successes have rallied calls to replace or avoid chemotherapy. Yet, even in this era of precision medicine and exciting advances, cytotoxic chemotherapies remain an essential component of lung cancer treatment. In the setting of locoregional disease, chemotherapy is the only systemic therapy thus far proven to enhance curability when combined with surgery or radiation. In the metastatic setting, chemotherapy can improve the length and quality of life in many patients. Chemotherapy remains the mainstay of care for individuals whose cancers with oncogenic drivers have acquired resistance to targeted agents. Chemotherapy also has the potential to modulate the immune system to enhance the effectiveness of immune checkpoint inhibitors. In this context, chemotherapy should be framed as a critical component of the armamentarium available for optimizing cancer care rather than an unfortunate anachronism. We examine the role of chemotherapy with precision medicine in the current care of patients with lung cancers, as well as opportunities for future integration in combinations with targeted agents, angiogenesis inhibitors, immunotherapies, and antibody drug conjugates.

introduction

The development of therapies for lung cancers covers three distinct eras. The cytotoxic chemotherapy era began more than 65 years ago [1, 2] and over the subsequent decades, a multitude of studies have demonstrated the effectiveness of chemotherapy in patients with lung cancers. The last 15 years have featured molecularly targeted therapies made possible by an extraordinary expansion of our understanding of the biology of lung cancers [3–5], a revolution in the tools available to molecularly characterize tumors [6, 7], and advances in targeted therapy development [8–13]. It is now a standard of care to obtain sufficient tissue from lesions suspected to be lung cancers to precisely determine both the pathologic cell type and genotype at diagnosis [14, 15]. The immunotherapy era of the last 5 years, emerging after several decades of disappointment in vaccines and other immunotherapies, has firmly established the proof-of-principle that the immune system can be leveraged to treat patients with lung cancers. The exceptional ability of immunotherapy to produce durable and perhaps permanent benefits [16] has resulted in unprecedentedly quick regulatory approval for these therapies [17–20] and widespread acceptance by both clinicians and patients while providing broad new avenues for drug development. Still, substantial work remains to better identify the determinants of response such that the use of immunotherapies may approximate the personalized use of molecularly targeted agents.

Together, these advances have substantially increased length and quality of life for patients with lung cancers such that average life spans are measured in years rather than months. Even in persons with no targetable oncogene in their tumors, the median survival from the time of diagnosis of metastatic disease is 2.1 years (and up to 4 years to patients with EGFR mutated or ALK rearranged lung adenocarcinomas) [14]. Molecularly targeted agents routinely demonstrate response rates of ∼70% [9, 21–24]; however, after a median period of 8–16 months, relapse occurs in almost all individuals. As resistance is inevitable to targeted therapies and progression-free survival (PFS) is measured in months, is the reasonable question to ask: how is multi-year survival achieved? The answer is the substantial advances of targeted therapies occur in addition to the routine use of intravenous chemotherapies given before, after, or concomitantly with targeted therapies. It is not ‘new therapies replacing chemotherapy’ but ‘new therapies in addition to chemotherapy’ that more accurately captures both the mode of our progress and the current treatment paradigm for the care of patients with lung cancers (Figure 1).

Figure 1.

Depiction of how chemotherapy plays a central role in the care of all persons with lung cancers.

Yet, the successes of molecularly targeted and immunotherapies have been used to denigrate the efficacy of intravenous chemotherapy and its ultimate contribution to improved survival in most patients with lung cancers. Chemotherapy is framed as something to be avoided, which can result in patients and providers alike declining the opportunity for potentially effective therapy. In a prospective series of patients in British Columbia, Canada, diagnosed with advanced non-small-cell lung cancer in 2014, only 55% of patients were even referred to medical oncology [25]. It is critical that oncologists participate in reframing the dialogue to highlight that success of novel agents in clinical trials compared with chemotherapy is not a zero-sum game. This progress should be heralded as the addition of therapies available in our continued effort to optimize outcomes for patients. Such discussion is particularly timely as several phase III studies examining the comparison of first-line chemotherapy versus immunotherapy (as monotherapy and as combination) are expected to read-out initial results soon. Further, to ensure our progress does not plateau, it is also worthwhile to continue to consider creative ways to combine chemotherapy with both molecularly targeted and immunotherapeutic agents.

chemotherapy works

chemotherapy can produce cures in early-stage lung cancers

Regardless of driver-mutation and PD-L1 status, cytotoxic chemotherapy remains the only form of systemic therapy proven to improve the chance of cure in early-stage non-small-cell lung cancers, when used in combination with surgery and/or radiation therapy. Adjuvant (postoperative) cisplatin-based chemotherapy has been repeatedly demonstrated to improve survival in completely resected stage IB–IIIA cancers [26–31]. The best estimates of the added benefit of chemotherapy compared with surgery alone by stage are derived from the LACE (Lung Adjuvant Cisplatin Evaluation) meta-analysis, with the number needed to treat to save one life being 33, 10, and 8 for stage IB, II, and III, respectively [32]. Neoadjuvant chemotherapy has an equivalent benefit to adjuvant chemotherapy in stages IB–IIIA non-small-cell lung cancers [33, 34], while permitting better drug tolerability, dose delivery [35], and assessment of interval response to chemotherapy both radiologically [36] and pathologically [37]. Although the decision whether or not to pursue perioperative chemotherapy should be personalized, the potential of chemotherapy to transform an individual's binary outcome from ‘not cured’ to ‘cured’ is profound, even if only a minority of patients receiving chemotherapy benefit.

For patients with bulky N2 disease or N3 non-small-cell lung cancers where radiation is the preferred definitive modality, several randomized studies have similarly shown that the addition of chemotherapy improves survival. The superiority of concurrent over sequential chemoradiation was demonstrated in a randomized trial showing a 6% absolute 5-year survival advantage in patients receiving concurrent therapy [38]. Pre-operative concurrent chemoradiotherapy is the standard of care for superior sulcus or Pancoast tumors [39]. Some institutions offer this approach with neoadjuvant chemoradiotherapy before resection of other stage III lung cancers, although this practice is debatable as the outcomes of randomized trials have revealed increased toxicity and no benefit over neoadjuvant chemotherapy alone [34, 40, 41].

Seen through this lens, chemotherapy represents a critical component of the care for nearly all patients with locoregional non-small-cell lung cancers and is able to achieve an end point that no other systemic therapy has proven: cure. To date, while the data showed adjuvant tyrosine kinase inhibitors can improve outcomes and delay relapse in patients with complete resections of EGFR-mutant lung cancers [42], prospective studies have not yet demonstrated the capacity of EGFR-targeted therapy to modify the fraction of patients who achieve cure. The NCI-sponsored ALCHEMIST initiative should answer the question of the role of adjuvant erlotinib in EGFR-mutant and crizotinib in ALK-positive lung cancers, after the completion of chemotherapy. An additional arm to study the T-cell checkpoint inhibitor nivolumab after chemotherapy is also planned [43]. The design of these studies, employing a targeted or checkpoint inhibitor after completion of standard chemotherapy and/or radiotherapy, emphasizes the role of cytotoxic chemotherapy as an essential part of the curative treatment for lung cancer.

To date, only chemotherapy can eradicate micrometastasis. But this is, of course, not to say that chemotherapy has achieved satisfactory outcomes in patients with early-stage lung cancers. Enormous improvement is still needed and it is certain the optimal perioperative treatment of lung cancers will continue to evolve. However, chemotherapy is likely to stay and both targeted therapies and immunotherapies will build upon rather than necessarily replace chemotherapy.

chemotherapy improves survival in advanced lung cancers

To properly capture the effects of chemotherapy in patients with advanced lung cancers, it is worthwhile to examine the outcomes that occur in the absence of chemotherapy. In randomized trials of chemotherapy versus best supportive care or placebo conducted in the 1970s and 1980s, the median survival in those not receiving chemotherapy ranged from 9 to 17 weeks and overall survival at 1 year hovered in the range of 10%–15% [44–47]. Fast-forward to today, modern chemotherapies have resulted in a substantial increase in 1-year survival: 51% in ECOG 4599 [48], 58% in PARAMOUNT [49], and 53%–54% in PointBreak [50]. Again, there remains substantial need for further improvement of course, but our ongoing progress is built upon a foundation of the role for chemotherapy and the benefits of chemotherapy alone can be important.

Trials of chemotherapy have demonstrated that a plurality, if not a majority, of patients benefit to some degree, but durable responses can also be seen. In a series of 364 patients with advanced lung adenocarcinomas from our institution treated with first-line platinum/pemetrexed-based chemotherapy from 2009 to 2011, 52 (14%, 95% confidence interval 11% to 19%) remained on therapy for longer than 1 year. This proportion (albeit small) of patients who do achieve long-term benefit from first-line chemotherapy may be relevant when considering what can also be achieved with anti-PD-1 monotherapy. We eagerly await the results of such prospective trials that are expected to have initial reportable results soon [e.g. KEYNOTE-024 (NCT02142738), KEYNOTE-042 (NCT02220894), and CheckMate-026 (NCT02041533)], which may well yield new options for patients with PD-L1 expressing NSCLCs. Recently updated data from KEYNOTE-001, for example, reported in patients with PD-L1 expression greater than or equal to in patients treated with first-line pembrolizumab, 54% were progression-free at 1 year [51]. And building upon recent promising data for dual immunotherapy combinations (durvalumab/tremelimumab, with provocative increases in response rates in PD-L1-negative patients relative to what would be expected with durvalumab monotherapy [52], as well as nivolumab/ipilimumab, with response rates of 39%–47% among all comers, 57% in those with PD-L1 ≥1%, and up to 92% in those with PD-L1 ≥50%) [53], trials of these combinations compared with chemotherapy and PD-1/PD-L1 monotherapy are actively ongoing [MYSTIC (NCT02453282), NEPTUNE (NCT02542293), CheckMate-227 (NCT02477826)]. Regardless of the outcomes of these studies, it is important to note that the conclusions should be interpreted neither as a repudiation of chemotherapy (if positive) nor disappointment in the role for immunotherapy (if negative). It is by using all potentially effective therapies together, either in sequence or concurrently, that we can most effectively continue to raise the survival curve for patients with advanced lung cancers [16].

In the era of targeted therapy, randomized trials comparing targeted agents with chemotherapy have also produced an intriguing observation that patients with oncogene-driven lung cancers have particularly high response rates to chemotherapy. In the IPASS trial comparing gefitinib with the combination of carboplatin and paclitaxel, the response rate to chemotherapy in patients with EGFR-mutant lung cancers was double that of EGFR mutation-negative patients (47%versus 24%) [9]. In the PROFILE 1014 and PROFILE 1007 trials comparing crizotinib with chemotherapy in patients with ALK fusion-positive lung cancers in the first- and second-line settings, respectively [22], response rates were 45% with cisplatin and pemetrexed in first line, and 20% with pemetrexed or docetaxel in second line [22, 54]. These observed chemotherapy activity is substantially higher than what was seen in earlier trials of the same agents in molecularly unselected patients [55]. Ongoing investigations are examining the sensitivity of various oncogene-driven lung cancers to chemotherapy, in particular pemetrexed for ALK fusions [56, 57], and more recently pemetrexed for RET [58] and ROS1 fusions [59, 60].

opportunities for future development of chemotherapy

It is commonly predicted that the benefit derived from cytotoxic chemotherapy has plateaued. This may be right for cytotoxic chemotherapy by itself, but we remain hopeful that combinations of chemotherapy with molecularly targeted or immune therapies may well be synergistic.

chemotherapy in combination with molecularly targeted therapies

In preclinical models, high pulsatile doses of gefitinib immediately before chemotherapy sensitized tumors to significantly greater anticancer effects of chemotherapy compared with standard continuous dosing of gefitinib [61]. When this observation was tested clinically, patients with EGFR wild-type lung cancers given pulsatile high dose of erlotinib 1500 mg daily for 2 days before carboplatin plus paclitaxel experienced improved response rate and survival compared when the same chemotherapy was given alone or when high dose of erlotinib was given for 2 days after chemotherapy [62]. In the FAST ACT trial, patients with chemotherapy intercalated with erlotinib had significantly longer PFS compared with patients who received chemotherapy alone [63]. This finding was confirmed in the FAST ACT-2 randomized phase III trial, where overall survival benefit was also seen in patients who received chemotherapy intercalated with erlotinib [64], a result that has never been achieved in multiple randomized trials comparing single-agent erlotinib and chemotherapy [65, 66]. Taken together, these results show that combining chemotherapy with EGFR-targeted therapy is safe, and support the hypothesis that combining, sequencing, or intercalating chemotherapy with EGFR tyrosine kinase inhibitors may be effective strategies to improve outcomes of patients with EGFR-mutant lung cancers. These strategies are being investigated in a follow-up to FAST ACT-2, evaluating gefitinib, chemotherapy, or the combination of the two (NCT02148380).

Advances in bioengineering have led to the emergence of a new class of agents, antibody drug conjugates, which combine cytotoxic chemotherapy and targeted agent into a single construct. Following the approval of ado-trastuzumab emtansine and brentuximab vedotin as new standards of care for patients with HER2-positive breast cancers and CD30-positive lymphomas, respectively [67–69], multiple antibody drug conjugates are now in development for patients with lung cancers. Ado-trastuzumab emtansine, currently in development for HER2 amplified or mutant lung cancers [70], leverages both the HER2 targeting properties of trastuzumab and the cytotoxic properties of emtansine, a maytansinoid chemotherapy, as a potentially new treatment for patients with HER2-amplified or mutant lung cancers [71]. Related to small cell lung cancers, a recent extraordinary example of the synergistic potential of targeted and cytotoxic therapy is rovalpituzumab tesirine (Rova-T), an antibody drug conjugate targeting protein expression of delta-like protein 3 (DLL3). In a phase I trial, Rova-T produced a confirmed objective response rate of 16% in patients with refractory extensive stage small cell lung cancers, and up to 31% among those with high DLL3 protein expression [72], a particularly important advance for this refractory disease.

chemotherapy in combination with immunotherapies

Although sometimes considered mutually antagonistic, the mechanisms of action of cytotoxic chemotherapy and immunotherapies may be far more intertwined and ultimately synergistic than initially thought [73]. Importantly, the relationship between efficacy of chemotherapy and anti-tumor immunity can be bidirectional. The presence of pre-existing tumor-infiltrating lymphocytes can substantially improve response to chemotherapy in a variety of solid tumors [74–76] and anti-tumor responses of chemotherapies are abated in the absence of components of the immune system in pre-clinical models [77, 78]. Intriguingly, there is increasing recognition of the ways in which responses to molecularly targeted agents may be, at least in part, immune-mediated [79, 80]. These observations have prompted provocative suggestions that many conventional chemotherapies and targeted therapies may well be conceptualized as immunotherapies [81, 82]. Regardless, it is beyond question that chemotherapy has the potential to modulate the tumor and its tumor microenvironment to favor anti-tumor effector responses [83–88]. Several demonstrated mechanisms include increased antigenicity by release and presentation of tumor-specific antigens [89] and up-regulation of major histocompatibility complex expression [90, 91], induction of signals that promote antigen-presentation by dendritic cells [92, 93], selective depletion of specific immunosuppressive cells in the tumor microenvironment such as myeloid-derived suppressor cells (MDSCs) and regulatory T cells (Tregs) [94–98], and enhancing effector T cells [99, 100].

The capacity of chemotherapy to productively modulate anti-tumor immunity has prompted multiple studies combining chemotherapy with immunotherapies. Several chemotherapies, including paclitaxel and cisplatin, have been demonstrated to synergize with cancer vaccines in pre-clinical [101, 102] and clinical models [103]. However, the most substantial recent effort has been directed at combining chemotherapy with T-cell checkpoint blockade. Anti-CTLA-4 and anti-PD-1/PD-L1 therapies have revolutionized the treatment of patients with a multitude of cancers, including lung cancers. But responses occur in only a minority of patients prompting eagerness to identify rationale combination regimens that effectively increase the frequency and durability of response. The first of these clinical studies examined the effect of combining cytotoxic chemotherapy with T-cell checkpoint inhibitors in patients with advanced NSCLCs [104] or SCLCs [105]. In both studies, patients were randomized to carboplatin and paclitaxel alone, carboplatin and paclitaxel plus ipilimumab (anti-CTLA-4) started concurrently with chemotherapy, or carboplatin and paclitaxel plus ipilimumab started in a phased fashion, starting after two cycles of chemotherapy. In patients with NSCLCs, PFS was significantly longer in the phased group compared with chemotherapy alone (HR 0.72, P = 0.05), while PFS in the concurrent treatment arm was numerically but not statistically significantly longer (HR = 0.81, P = 0.13) [104]. Similar results were seen in patients with SCLC, where the addition of phased ipilimumab significantly increased PFS but concurrent ipilimumab did not (HR 0.64, P = 0.03, HR 0.93, P = 0.37, respectively) [105]. These preliminary results have prompted phase III studies in both NSCLC (NCT02279732, NCT01285609) and SCLC (NCT01450761), but results have not yet been reported.

Understandably, more recent studies have focused on combinations of chemotherapy with anti-PD-1 or PD-L1 agents. In CheckMate-012, 56 patients with advanced NSCLCs were treated with nivolumab in combination with a variety of chemotherapy regimens (carboplatin or cisplatin + pemetrexed, gemcitabine, or paclitaxel) [106]. The overall response rate ranged from 33% to 47% with the median PFS ranging from 5 to 7 months, which was not obviously better than what might have been expected from chemotherapy alone. There was also notable toxicity in which 45% patient experienced treatment-related grade 3 or 4 toxicities. Of note, most patients in this study received a higher dose of nivolumab than is currently FDA approved (10 versus 3 mg/kg), which may have contributed to the increased toxicity seen.

Data presented at ASCO 2016 provided more encouraging, but still preliminary, results for combinations of chemotherapy with PD-1 or PD-L1 blockade. In KEYNOTE-021, the response rate to pembrolizumab/carboplatin + paclitaxel was 52% (n = 25), up to 70% in those treated with pembrolizumab/carboplatin + pemetrexed (n = 24) [107], and treatment was a bit better tolerated, with 0–1 patients discontinuing therapy due to toxicity in these arms and no treatment-related deaths. Additionally, in a study of atezolizumab combined with various chemotherapies [carboplatin + paclitaxel (n = 8), pemetrexed (n = 14), or albumin-bound paclitaxel (n = 15)], response rates ranged from 60% to 75% [108]. The combination of anti-PD-L1 therapy with albumin-bound paclitaxel is theorized to be particularly attractive due to the lack of need for concurrent steroid administration, but the chemotherapy regimen that optimally synergizes with T-cell checkpoint blockade remains unknown. Importantly, in both of these studies, the sample sizes are small and follow-up remains short. Further data are expected soon to better clarify whether the combination of chemotherapy with immunotherapy exceeds the benefit of either treatment alone or which specific combination of therapies may be most effective. Several phase III studies are now ongoing to answer this question (NCT02578680, NCT02367781, NCT02367794, NCT02366143, NCT02477826).

Beyond combinations of cytotoxic chemotherapy and T-cell checkpoint blockade, the question has also been raised as to whether the sequence of each regimen may impact response. The response rate to anti-PD-1 therapies when used initially is slightly higher numerically (23%–25% [20, 109]) compared with after initial chemotherapy (18%–20% [18, 19, 110]), but it is not clear that this small change in response rate is attributable to prior chemotherapy. Notably, responses to anti-PD-1 therapies have been observed in patients who have been previously pre-treated with chemotherapy, even in individuals who have received up to 7 prior lines of chemotherapy [20, 111]. In patients with squamous cell lung cancers treated with nivolumab as third-, fourth-, or fifth-line treatment, there are no major differences in response rate (10%, 17%, or 17%, respectively) [112].

In sum, although pre-clinical data provide a good rationale for combining cytotoxic chemotherapy with immunotherapy, there are no firm data yet to demonstrate a distinct advantage to administering an immune checkpoint inhibitor before, concomitantly with, or after chemotherapy. Multiple efforts are ongoing to examine this critical issue. In the meantime, available data demonstrate that prior chemotherapy, even multiple courses and regimens, does not appear to significantly abate or augment response rates to anti-PD-1 or anti-PD-L1 agents and it is reasonable to consider these immunotherapy agents at any point in the course of treatment of patients with lung cancers.

conclusion

Chemotherapy works. In patients with locoregional disease, chemotherapy can improve curability when combined with surgery or radiation. It is the only systemic therapy that has been proven to enhance curability. In the setting of metastatic disease, chemotherapy has profoundly improved both length and quality of life. Chemotherapy works better in patients with oncogenic drivers, can be safely combined with molecularly targeted agents to improve outcomes, and should be routinely used when resistance to molecularly targeted agents arises. It appears that chemotherapy may usefully synergize with immunotherapies, although much work remains to understand how best to link and integrate all effective modalities. Chemotherapy today is a foundational piece of treatment for nearly all patients with lung cancers and will remain an indispensable contributor to our efforts to improve the care of all persons with these illnesses.

funding

This research was funded in part by the National Cancer Institute at the National Institutes of Health (P30 CA008748).

disclosure

MDH has received consulting fees from Genentech/Roche, Bristol-Myers Squibb, Merck, AstraZeneca, Neon, Inovio Pharmaceuticals, Alexion; BTL has received consulting fees from Biosceptre International; JEC has received consulting fees from Genentech/Roche, AstraZeneca, and Clovis Oncology; MGK has received consulting fees from Ariad, AstraZeneca, and Genentech/Roche.