The impact of zoledronic acid therapy in survival of lung cancer patients with bone metastasis

Abstract

Bone metastases occur in 20–40% of patients with lung cancer. Recent studies demonstrate a direct antiproliferative effect of 3rd generation bisphosphonates (BPs) on lung tumors, which may influence the survival. Therefore, we examined the clinical impact of zoledronic acid (ZOL; Zometa®), a 3rd generation BP, with a focus on the survival, time to progression and pain effect in lung cancer patients with bone metastases. Lung cancer patients (n = 144, Stage IV) with evidence of metastasis bone scan were included. Eighty-seven of 144 experienced bone pain and received ZOL, 4 mg i.v. every 21 days (Group A), whereas the other 57 patients received no ZOL (Group B). All patients were treated with a combination chemotherapy consisted of docetaxel 100 mg/m² and carboplatin AUC = 6. It was found that Group A had a statistically significant longer survival (p < 0.01) when compared to Group B. A statistically significant positive correlation was found between the number of cycles of therapy with ZOL and total patient survival (p < 0.01, Pearson correlation) and time to progression (p < 0.01). Pain effect of ZOL had no significant difference between the 2 groups of patients (p > 0.05). Urine N-telopeptide of type I collagen (NTx) levels decreased in patients with NTx ≤ 29 nM BCE/mM creatinine at baseline after treatment with ZOL. The results of our study suggest that the addition of ZOL increases overall survival in lung cancer patients with bone metastases. The longer period of receiving ZOL, the better effect on survival and time to progression. © 2009 UICC

Lung cancer is the leading cause of death from cancer worldwide. Approximately 80% of lung cancers can be histologically classified as non-small-cell lung cancer (NSCLC). The majority of patients present with locally advanced (37%) or metastatic (38%) disease at the time of diagnosis.1 It has been estimated that ∼30–65% of patients with metastatic lung cancer will develop bone metastases.2, 3 Bone metastases cause considerable skeletal morbidity, including bone pain, pathologic fractures, spinal cord compression and hypercalcemia of malignancy.2 These skeletal-related events (SREs) are the result of the resorption of mineralized bone by osteoclasts. The management of skeletal complications is typically a multimodal endeavor involving surgery, radiation therapy, analgesics and more recently the administration of bisphosphonates (BPs).4 BPs have been extensively used in the treatment and prevention or palliation of skeletal complications associated with osteolytic lesions in patients with breast cancer,5 multiple myeloma6 and more recently in other solid tumors, such as lung cancer.

BPs act on bone cells such as osteoclasts and are generally used to treat lytic bone lesions caused by malignancies or bone resorption disorders such as osteoporosis. All BPs are characterized by a phosphorus–carbon–phosphorus (P–C–P)-containing central structure, which promotes their binding to the mineralized bone matrix, and a variable R chain, which determines the relative potency, side effects and probably also the precise mechanism of action. After administration, BPs bind avidly to exposed bone mineral around resorbing osteoclasts leading to very high local concentrations of BP in the resorption lacunae (up to 1,000 μM). On release from the bone surface, BPs are internalized by the osteoclast, where they cause disruption of the biochemical processes involved in bone resorption.7 BPs also cause osteoclast apoptosis, with the appearance of distinctive changes in cell and nuclear morphology. Although the molecular targets responsible for promoting this apoptosis are unknown, the BPs have recently been shown to inhibit enzymes of the mevalonate pathway, which are ultimately responsible for events that lead to the posttranslational modification of GTP-binding proteins such as Ras.8 Moreover, recent studies suggest that nitrogen-containing BPs may have direct apoptotic effects on tumor cells. There is an expanding database of preclinical evidence that BPs, especially the newer, 3rd generation, nitrogen-containing BPs, such as zoledronic acid (ZOL), can inhibit the proliferation and induce apoptosis in a broad range of human cancer cells,9 including cell lines derived from both small-cell lung cancer (SCLC) and NSCLC human tumors.10, 11

On the other hand, biochemical markers of bone metabolism, which reflect both the formation and resorption of bone,12 can provide valuable insight into tumor and bone interactions and the effects of therapy on this dynamic process. Metastatic bone disease is typically associated with a marked increase in bone resorption.13 Clinical studies suggest that bone resorption markers, particularly the N-telopeptide of type I collagen (NTx), are associated with the presence and extent of metastases,14, 15 prognosis16 and response to treatment.17

In our study, we examined the clinical impact of ZOL with a focus on the survival, time to progression and pain effect on lung cancer patients with bone metastases.

Material and methods

Patients, treatment and biochemical determinations

NSCLC patients (n = 144), Stage IV, median age 62 (95% CI: 59.5–64.4), with positive bone scan and 0–1 Eastern Cooperative Oncology group (ECOG) performance status, were consecutively recruited. Patients were divided into 2 groups according to their bone pain symptoms. Eighty-seven of 144 (Group A) experienced bone pain and the rest 57 (Group B) were asymptomatic. Twenty-five of 87 patients of Group A and 12 of 57 patients of Group B belong to ECOG PS 0. All patients had adequate renal and liver function and were treated with a combined chemotherapy (CHT) consisted of docetaxel 100 mg/m² and carboplatin AUC = 6/Q 4 weeks up to 8 cycles (according to clinical experience in our department).

Additionally, symptomatic bone pain patients (Group A) received ZOL, 4 mg i.v., as a single intravenous infusion in no less than 15 min18–20 initially every 28 days—during CHT—and every 21 days after the completion of CHT—as maintenance therapy—till the progression of the disease or the significant decline in performance status of the patients.

In Group A, urine markers of bone metabolism, NTx, calcium as well as creatinine in urine were monitored before treatment, in the 3rd, 6th, 9th and 12th cycles of therapy. The correlation between these bone resoption markers and risks for clinical outcomes were also estimated. Pain was assessed in both groups using a pain score measured on a scale of 0–1021 in each clinical visit. Seventy-eight patients of Group A scored up to 4 in the pain scale, 8 patients between 4 and 6 and only 1 patient above 8.

Patients of Group A were recommended to have a dental examination and were advised on appropriate measures for reducing the risk of jaw osteonecrosis.22 Patients in Group A were also administered an oral calcium supplement of 500 mg and 400 IU vitamin D daily.

Patients who respond to CHT received radiotherapy to the primary site (50 Gy) between the 2nd and the 3rd cycle. Base line investigations before treatment initiation included: complete blood count with differential and platelet count, complete biochemical profile, electrocardiograph (ECG), chest films, computed tomography (CT) of the chest and abdomen, CT of the brain (or magnetic resonance imaging, if necessary), isotope bone scan and bronchoscopy. Extra X-rays or bone MRIs were performed where it was needed. All patients had only bone metastasis, and no other site of metastasis was observed at recruitment. The patients were monitored weekly with full blood count as the onset on nadir occurred as early as Day 6 (due to taxane) and to be evaluated for the need of supportive treatment. Biochemical profile was performed before every new CHT cycle, as well as chest radiographs to estimate response to therapy. After the completion of CHT, we restaged our patients.

Palliative radiotherapy and hematopoietic growth factors were used as appropriate. In specific patient situation, additional investigation and/or palliative treatment were used as necessary. Patients experiencing early progression were treated with salvage therapy, according to their clinical situation. The protocol was approved by the Hospital Ethical Committee. Informed consent was obtained from all screened patients.

In 6 of 87 patients, there was evidence of renal deterioration during treatment, but within 10% of baseline, so the BP therapy was continued.19 One patient was withheld from the protocol because of persistent renal deterioration, and 4 patients were also withheld during the study because of jaw osteonecrosis. However, all these patients were included in the statistical evaluation of the survival.

Statistical analysis

The primary efficacy analysis included time to progression of overall disease and survival. Other secondary efficacy variables included bone lesion response, change from baseline in bone pain inventory (BPI) composite pain score and changes from baseline in biochemical markers of bone resorption.

The Kaplan–Meier method and the log-rank test were used to estimate differences in survival and time to progression between the Groups A and B. For time to progression of overall disease, death due to progression of disease was counted as an event. Two-tailed Pearson correlation was used to compare the number of chemotherapies and time to progression and survival.

Cox regression survival analysis was performed to find whether any differences between the Groups A and B existed in survival and time to progression regarding type of bone metastasis (osteolytic, osteoblastic or mixed), sex, radiotherapy and line (1st, 2nd or 3rd) of CHT.

Results

The patient response rates are shown in Table I. Median survival was 578 days (95% CI: 454–701.8) for Group A and was statistically significant higher in comparison to Group B with median survival 384 days (95% CI: 368–399.6), p < 0.001 (Fig. 1).

Figure 1. Kaplan–Meier method and log-rank test. Median survival for Group A was higher in a statistically significant manner in comparison to Group B median survival.

Table I. Patient Response Rates

Patients groups	Partial response	Stable disease	Progressive disease
Group A	28 (28.7%)	37 (42.5%)	25 (28.7%)
Group B	9 (15.7%)	25 (43.8%)	23 (40.3%)

Median time to progression was 265 days (95% CI: 240.5–289) for Group A and was statistically significant higher in comparison to 150 days (95% CI: 56–244) for Group B, p < 0.001 (Fig. 2).

Figure 2. Kaplan–Meier method and log-rank test. Median time to progression for Group A was higher in a statistically significant manner in comparison to Group B median time to progression.

The median number of CHT cycles that was given in Group A was 8 (95% CI: 7.9–10.15) and in Group B was 7 (95% CI: 6.4–7.6). Additionally in Group A, the median number of ZOL cycles was 9 (95% CI: 6.7–11), during CHT and maintenance.

A statistically significant positive correlation was found between the number of cycles of therapy with ZOL and total patient survival (p < 0.01 Pearson correlation) and time to progression (p < 0.01) (Fig. 3).

Figure 3. Pearson correlation. Increased numbers of cycles of therapy with zoledronic acid augment total patient survival.

There was no statistical significant difference between the 2 groups of patients in relation to the pain effect of ZOL in comparison to baseline (p > 0.05). As far as SRE concerned, 3 of 57 patients of Group B (5.2%) and 4 of 87 of Group A (4.5%) experienced SRE during 1–4 and 2–4 months, respectively, after patients enrollment in the study for the 2 groups.

The mean of urine N-telopeptide of type I bone collagen equivalent (NTx nM BCE) levels in Group A, which received ZOL reduced from 447.29 before treatment in 204.85 after 3 months and was preserved, with little fluctuations, stable till the 12th month after the beginning of therapy (Fig. 4). The mean levels of NTx appeared with fluctuations among the initial and every 3 months measurements without statistical significant difference (Fig. 5). Nevertheless, when we divided patients of Group A into 2 subgroups according to their NTx levels, the subgroup A₁ with NTx ≤ 29 nM BCE/mM creatinine and the subgroup A₂ with NTx > 30 nM BCE/mM creatinine, we found that NTx in subgroup A₁ continuously decreased, reaching the lowest levels in the 9th month—although not statistically significant—followed by a little increase in the 12th month. On the other hand, NTx in subgroup A₂ presented consecutive fluctuations before and during the treatment (Fig. 6). Moreover, a statistically positive correlation (p < 0.05) was observed between the response of patients in CHT and NTx levels (Fig. 7). A statistically significant negative correlation exists between NTx (nM BCE levels) and time to progression in the 9th and 12th month measurements (Fig. 8).

Figure 4. Mean of NTx (nM BCE) before treatment and in 3rd, 6th, 9th and 12th cycle of therapy.

Figure 5. Mean levels of NTx nM BCE/mM creatinine before treatment and in 3rd, 6th, 9th and 12th cycle of therapy.

Figure 6. Mean levels of NTx before treatment and in 3rd, 6th, 9th and 12th cycle of therapy. Subgroup A₁: NTx ≤ 29 nM BCE/mM creatinine, Subgroup A₂: >30 nM BCE/mM creatinine.

Figure 7. Mean levels of NTx before treatment and in 3rd, 6th, 9th and 12th cycle of therapy. Subgroups were defined relatively to the response of patients to chemotherapy.

Figure 8. Correlation between NTx nM BCE levels and time to progression.

In calcium urine levels, there was a gradual reduction between the initial (2.074) and the levels after 6 months (1.25). This difference was statistically significant (p = 0.02) (Fig. 9).

Figure 9. Mean calcium urine levels (mM/mM creatinine) before treatment and in 3rd, 6th, 9th and 12th cycle of therapy (paired samples test).

No statistically significant differences between Groups A and B were found in survival and time to progression regarding type of bone metastasis (osteolytic, osteoblastic or mixed), sex, radiotherapy and 1st, 2nd or 3rd CHT line.

Toxicity and adverse events

Adverse events associated with ZOL were generally well tolerated and manageable and are shown in Table II. No significant changes were detected in laboratory parameters including blood chemistry, calcium or kidney function. Moreover, in our study, there was no statistical difference in the toxicity from CHT between patients who received NBPs and patients who received no NBPs.

Table II. Adverse Events Associated With Zoledronic Acid

Adverse events	No. of patients	%
Flu-like symptoms	13	14
Bone pain	10	11
GI complaints	15	17
Lower limb edema	3	3
Jaws osteonecrosis	4	5
Increase of creatinine ≤1 mg/dl	4	5

Discussion

ZOL demonstrates clinical benefit not only for patients with multiple myeloma and bone metastases from breast and prostate cancer but also for patients with lung cancer. Except better treatment of bone metastases and a consistent reduction of skeletal morbidity, which justify the higher survival, there is evidence that ZOL is shown to exert synergistic antitumor activity when combined with other anticancer agents.23–25 Indeed, in our study, there is a survival benefit of almost 6 months in patients who received ZOL compared to patients who received no ZOL (19 vs. 12.8 months or more accurately 578 vs. 384 days). Regarding time to progression, this benefit was close to 4 months (8.8 vs. 5 months or 265 vs. 150 days). The time to progression and the survival benefit appear to be due to Zol administration regardless of bone metastasis type (osteolytic, osteoblastic or mixed), sex, radiotherapy and CHT line (1st, 2nd or 3rd). Although survival and time to progression were higher in Group A in comparison to Group B, no statistical significant difference was found in bone pain scale between the 2 groups. A possible explanation could be that the majority of our patients had a score in bone pain scale up to 3. Moreover, CHT, which was given at the same time with NBPs, could have a favorable impact in bone pain. Furthermore, the longer period of receiving ZOL, the better effect on survival and time to progression.

The measurement of biochemical markers of bone metabolism seems to have prognostic significance. Patients who developed or maintained high NTx at baseline or during the course of ZOL therapy were at higher risk for deterioration and poor clinical outcome. Although these analyses cannot directly address a causal link between decreases in bone metabolism and clinical outcome, they do suggest that reducing bone turnover should have a positive effect by delaying progression of bone lesions and possibly improving survival.

Urinary NTx was characterized as low (NTx < 60 nM BCE/mM creatinine) or high (NTx > 0 nM BCE/mM creatinine). The cutoff values for NTx were chosen to reflect the approximate Upper Limit of Normal (ULN). However, the normal range for urinary NTx varies according to age, sex and endocrine function, with ∼50 nM/mM creatinine being the approximate ULN in young healthy adults and ∼100 nM/mM creatinine being the approximate ULN after menopause in women or during androgen deprivation therapy in men.26 Urinary NTx was normalized to the level of urinary creatinine.

Arbitrarily, we divided patients of Group A into those with high levels (>30 nM BCE/mM creatinine) and those with low levels (≤29 nM BCE/mM creatinine) of NTx, based on observed correlation between NTx levels and patient survival. Patients with initial low NTx levels had better TTP and survival, whereas patients with initial high urine NTx levels were found to be associated with worse survival and early TTP. This finding is in accordance with results from other studies, which showed correlation between initial high urine NTx levels and a 4- to 6-fold increased risk for skeletal complications and disease progression.27 The gradual reduction of calcium urine levels and particularly its significant decrease (p = 0.02) after 6 months ZOL treatment, which is well correlated to NTx, may well be attributed to the decrease of osteolysis.

Adverse events commonly associated with NBPs were generally manageable except jaw osteonecrosis—probably because of poor dental condition especially of elderly people in our country—and for this reason, we order panoramic X-ray in our patients. Safety data have been reported from a large database of Phase III clinical trial of ZOL enrolling >3,000 patients with bone metastases treated for up to 2 years.28, 29 Several retrospective studies have also reported the safety of ZOL treatment beyond 2 years.30 Ali et al.31 reported the safety of ZOL in patients with either multiple myeloma or breast cancer who received treatment for a mean of 3.6 years.

In conclusion, the addition of BPs seems to increase overall survival in lung cancer patients with bone metastases, and the prolongation of NBPs administration is correlated with longer survival and time to progression. Furthermore, initial NTx to creatinine ratios seem to have a negative correlation with time to progression and patient survival, whereas response to CHT and survival is correlated with lower NTx to creatinine ratio levels (possibility to use as prognostic/predictive factor to choose patients for NBPs treatment). Further studies are needed to support the potential usefulness of NBPs as a therapeutic agent against lung cancer.